input
stringlengths
8.62k
18k
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> was awarded a second Navy Cross. Puller led American Marines and Nicaraguan National Guardsmen into battle against Sandinista rebels in the last major engagement of the Sandino Rebellion near El Sauce on December 26, 1932. After his service in Nicaragua, Puller was assigned to the Marine detachment at the American Legation in Beijing, China, commanding a unit of China Marines. He then went on to serve aboard , a cruiser in the Asiatic Fleet, which was commanded by then-Captain Chester W. Nimitz. Puller returned to the States in June 1936 as an instructor at The Basic School in Philadelphia, where he trained Ben Robertshaw, Pappy Boyington, and Lew Walt. In May 1939, he returned to the Augusta as commander of the on-board Marine detachment, and then back to China, disembarking in Shanghai in May 1940 to serve as the executive officer and commanding officer of 2nd Battalion, 4th Marines (2/4) until August 1941. Major Puller returned to the U.S. on August 28, 1941. After a short leave, he was given command of 1st Battalion, 7th Marines (1/7) of the 1st Marine Division, stationed at New River, North Carolina (later Camp Lejeune). World War II Early in the Pacific theater, the 7th Marines formed the nucleus of the newly created 3rd Marine Brigade and arrived to defend Samoa on May 8, 1942. Later they were redeployed from the brigade and on September 4, 1942, they left Samoa and rejoined the 1st Marine Division at Guadalcanal on September 18, 1942. Soon after arriving on Guadalcanal, Lt. Col. Puller led his battalion in a fierce action along the Matanikau, in which Puller's quick thinking saved three of his companies from annihilation. In the action, these companies were surrounded and cut off by a larger Japanese force. Puller ran to the shore, signaled a United States Navy destroyer, the USS Ballard, and then Puller directed the destroyer to provide fire support while landing craft rescued his Marines from their precarious position. U.S. Coast Guard Signalman First Class Douglas Albert Munro—Officer-in-Charge of the group of landing craft, was killed while providing covering fire from his landing craft for the Marines as they evacuated the beach and was posthumously awarded the Medal of Honor for the action, to date the only Coast Guardsman to receive the decoration. Puller, for his actions, was awarded the Bronze Star Medal with Combat "V". Later on Guadalcanal, Puller was awarded his third Navy Cross, in what was later known as the "Battle for Henderson Field". Puller commanded 1st Battalion 7th Marines (1/7), one of two American infantry units defending the airfield against a regiment-strength Japanese force. The 3rd Battalion of the U.S. Army's 164th Infantry Regiment (3/164) fought alongside the Marines. In a firefight on the night of October 24–25, 1942, lasting about three hours, 1/7 and 3/164 sustained 70 casualties; the Japanese force suffered over 1,400 killed in action, and the Americans held the airfield. He nominated two of his men (one being Sgt. John Basilone) for Medals of Honor. He was wounded himself on November 9. Puller was then made executive officer of the 7th Marine Regiment. While serving in this capacity at the Battle of Cape Gloucester, Puller was awarded his fourth Navy Cross for overall performance of duty between December 26, 1943, and January 19, 1944. During this time, when the battalion commanders of 3rd Battalion, 7th Marines (3/7) and later, 3rd Battalion, 5th Marines (3/5), were under heavy machine gun and mortar fire, he expertly reorganized the battalion and led the successful attack against heavily fortified Japanese defensive positions. He was promoted to colonel effective </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> February 1, 1944, and by the end of the month had been named commander of the 1st Marine Regiment. In September and October 1944, Puller led the 1st Marine Regiment into the protracted battle on Peleliu, one of the bloodiest battles in Marine Corps history, and received his first of two Legion of Merit awards. The 1st Marines under Puller's command lost 1,749 out of approximately 3,000 men, but these losses did not stop Puller from ordering frontal assaults against the well-entrenched enemy. The corps commander had to order the 1st Marine Division commanding general to pull the annihilated 1st Marine Regiment out of the line. During the summer of 1944, Puller's younger brother, Samuel D. Puller, the executive officer of the 4th Marine Regiment, was killed by an enemy sniper on Guam. Puller returned to the United States in November 1944, was named executive officer of the Infantry Training Regiment at Camp Lejeune and, two weeks later, commanding officer. After the war, he was made director of the 8th Reserve District at New Orleans, and later commanded the Marine barracks at Pearl Harbor. Korean War At the outbreak of the Korean War, Puller was once again assigned as commander of the First Marine Regiment. He participated in the landing at Inchon on September 15, 1950, and was awarded the Silver Star Medal. For leadership from September 15 through November 2, he was awarded his second Legion of Merit. He was awarded the Distinguished Service Cross from the U.S. Army for heroism in action from November 29 to December 4, and his fifth Navy Cross for heroism during December 5–10, 1950, at the Battle of Chosin Reservoir. It was during that battle that he said the famous line, "We've been looking for the enemy for some time now. We've finally found him. We're surrounded. That simplifies things." In January 1951, Puller was promoted to brigadier general and was assigned duty as assistant division commander (ADC) of the 1st Marine Division. On February 24, however, his immediate superior, Major General O.P. Smith, was hastily transferred to command IX Corps when its Army commander, Major General Bryant Moore, died. Smith's transfer left Puller temporarily in command of the 1st Marine Division until sometime in March. He completed his tour of duty as assistant commander and left for the United States on May 20, 1951. He took command of the 3rd Marine Division at Camp Pendleton, California until January 1952, and then was assistant commander of the division until June 1952. He then took over Troop Training Unit Pacific at Coronado, California. In September 1953, he was promoted to major general. Post-Korean War In July 1954, Puller took command of the 2nd Marine Division at Camp Lejeune, North Carolina until February 1955 when he became Deputy Camp Commander. He suffered a stroke, and was retired by the Marine Corps on November 1, 1955, with a tombstone promotion to lieutenant general. Regarding his nickname, in a handwritten addition to a typed 22 November 1954 letter to Major Frank C. Sheppard, Puller wrote, "I agree with you 100%. I had done a little soldiering previous to Guadalcanal and had been called a lot of names, but why 'Chesty'? Especially the steel part??" Relations Puller's son, Lewis Burwell Puller, Jr. (generally known as Lewis Puller), served as a Marine lieutenant in the Vietnam War. While serving with 2nd Battalion, 1st Marines (2/1), Lewis Jr. was severely wounded by a mine explosion, losing both legs and parts of his hands. Lieutenant General Puller broke down sobbing at seeing his son for the first time in the hospital. </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Lewis Jr. won a 1992 Pulitzer Prize for his autobiography, Fortunate Son: The Healing of a Vietnam Vet. He committed suicide in 1994. Puller was father-in-law to Colonel William H. Dabney, USMC (retired), a Virginia Military Institute (VMI) graduate, who was the commanding officer (then Captain) of two heavily reinforced rifle companies of the 3rd Battalion, 26th Marines (3/26) from January 21 to April 14, 1968, in Vietnam. During the entire period, Colonel Dabney's force stubbornly defended Hill 881 South, a regional outpost vital to the defense of the Khe Sanh Combat Base during the 77-day siege at the Battle of Khe Sanh. Dabney was recommended for the Navy Cross for his actions on Hill 881 South, but his battalion executive officer's helicopter carrying the recommendation papers crashed and the papers were lost. It was not until April 15, 2005, that Colonel Dabney received the Navy Cross during an award ceremony at Virginia Military Institute. Puller was a distant cousin to the famous U.S. Army General George S. Patton. He was an Episcopalian and parishioner of Christ Church Parish in Saluda and is buried in the historic cemetery there next to his wife, Virginia Montague Evans. Decorations and awards Puller received the second-highest U.S. military award five times (one of only two persons so honored): five Navy Crosses and one U.S. Army Distinguished Service Cross. He was the second of two U.S. servicemen (after U.S. Navy submarine commander Roy Milton Davenport) to ever receive five Navy Crosses. Puller's military awards include: First Navy Cross citation Citation: For distinguished service in the line of his profession while commanding a Nicaraguan National Guard patrol. First Lieutenant Lewis B. Puller, United States Marine Corps, successfully led his forces into five successful engagements against superior numbers of armed bandit forces; namely, at LaVirgen on 16 February 1930, at Los Cedros on 6 June 1930, at Moncotal on 22 July 1930, at Guapinol on 25 July 1930, and at Malacate on 19 August 1930, with the result that the bandits were in each engagement completely routed with losses of nine killed and many wounded. By his intelligent and forceful leadership without thought of his own personal safety, by great physical exertion and by suffering many hardships, Lieutenant Puller surmounted all obstacles and dealt five successive and severe blows against organized banditry in the Republic of Nicaragua. Second Navy Cross citation Citation: First Lieutenant Lewis B. Puller, United States Marine Corps (Captain, Guardia Nacional de Nicaragua) performed exceptionally meritorious service in a duty of great responsibility while in command of a Guardia Patrol from 20 September to 1 October 1932. Lieutenant Puller and his command of forty Guardia and Gunnery Sergeant William A. Lee, United States Marine Corps, serving as a First Lieutenant in the Guardia, penetrated the isolated mountainous bandit territory for a distance of from eighty to one hundred miles north of Jinotega, his nearest base. This patrol was ambushed on 26 September 1932, at a point northeast of Mount Kilambe by an insurgent force of one hundred fifty in a well-prepared position armed with not less than seven automatic weapons and various classes of small arms and well-supplied with ammunition. Early in the combat, Gunnery Sergeant Lee, the Second in Command, was seriously wounded and reported as dead. The Guardia immediately behind Lieutenant Puller in the point was killed by the first burst of fire, Lieutenant Puller, with great courage, coolness and display of military judgment, so directed the fire and movement of his men that the enemy were driven first from the high ground on the right of his position, and then </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> by a flanking movement forced from the high ground to the left and finally were scattered in confusion with a loss of ten killed and many wounded by the persistent and well-directed attack of the patrol. The numerous casualties suffered by the enemy and the Guardia losses of two killed and four wounded are indicative of the severity of the enemy resistance. This signal victory in jungle country, with no lines of communication and a hundred miles from any supporting force, was largely due to the indomitable courage and persistence of the patrol commander. Returning with the wounded to Jinotega, the patrol was ambushed twice by superior forces on 30 September. On both of the occasions the enemy was dispersed with severe losses. Third Navy Cross citation Citation: For extraordinary heroism as Commanding Officer of the First Battalion, Seventh Marines, First Marine Division, during the action against enemy Japanese forces on Guadalcanal, Solomon Islands, on the night of 24 to 25 October 1942. While Lieutenant Colonel Puller's battalion was holding a mile-long front in a heavy downpour of rain, a Japanese force, superior in number, launched a vigorous assault against that position of the line which passed through a dense jungle. Courageously withstanding the enemy's desperate and determined attacks, Lieutenant Colonel Puller not only held his battalion to its position until reinforcements arrived three hours later, but also effectively commanded the augmented force until late in the afternoon of the next day. By his tireless devotion to duty and cool judgment under fire, he prevented a hostile penetration of our lines and was largely responsible for the successful defense of the sector assigned to his troops. Fourth Navy Cross citation Citation: For extraordinary heroism as Executive Officer of the Seventh Marines, First Marine Division, serving with the Sixth United States Army, in combat against enemy Japanese forces at Cape Gloucester, New Britain, from 26 December 1943 to 19 January 1944. Assigned temporary command of the Third Battalion, Seventh Marines, from 4 to 9 January, Lieutenant Colonel Puller quickly reorganized and advanced his unit, effecting the seizure of the objective without delay. Assuming additional duty in command of the Third Battalion, Fifth Marines, from 7 to 8 January, after the commanding officer and executive officer had been wounded, Lieutenant Colonel Puller unhesitatingly exposed himself to rifle, machine-gun and mortar fire from strongly entrenched Japanese positions to move from company to company in his front lines, reorganizing and maintaining a critical position along a fire-swept ridge. His forceful leadership and gallant fighting spirit under the most hazardous conditions were contributing factors in the defeat of the enemy during this campaign and in keeping with the highest traditions of the United States Naval Service. Fifth Navy Cross citation Citation For extraordinary heroism as Commanding Officer of the First Marines, First Marine Division (Reinforced), in action against aggressor forces in the vicinity of Koto-ri, Korea, from 5 to 10 December 1950. Fighting continuously in sub-zero weather against a vastly outnumbering hostile force, Colonel Puller drove off repeated and fanatical enemy attacks upon his Regimental defense sector and supply points. Although the area was frequently covered by grazing machine-gun fire and intense artillery and mortar fire, he coolly moved along his troops to insure their correct tactical employment, reinforced the lines as the situation demanded, and successfully defended the perimeter, keeping open the main supply routes for the movement of the Division. During the attack from Koto-ri to Hungnam, he expertly utilized his Regiment as the Division rear guard, repelling two fierce enemy assaults which severely threatened the security of the unit, and personally supervised </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> the care and prompt evacuation of all casualties. By his unflagging determination, he served to inspire his men to heroic efforts in defense of their positions and assured the safety of much valuable equipment which would otherwise have been lost to the enemy. His skilled leadership, superb courage and valiant devotion to duty in the face of overwhelming odds reflect the highest credit upon Colonel Puller and the United States Naval Service. Distinguished Service Cross citation Citation: The President of the United States of America, under the provisions of the Act of Congress approved July 9, 1918, takes pleasure in presenting the Distinguished Service Cross to Colonel Lewis B. "Chesty" Puller (MCSN: 0-3158), United States Marine Corps, for extraordinary heroism in connection with military operations against an armed enemy of the United Nations while serving as Commanding Officer, First Marines, FIRST Marine Division (Reinforced), in action against enemy aggressor forces in the vicinity of the Chosin Reservoir, Korea, during the period 29 November to 4 December 1950. Colonel Puller's actions contributed materially to the breakthrough of the First Marine Regiment in the Chosin Reservoir area and are in keeping with the highest traditions of the military service. Silver Star citation Citation: The President of the United States of America, authorized by Act of Congress July 9, 1918, takes pleasure in presenting the Silver Star (Army Award) to Colonel Lewis B. "Chesty" Puller (MCSN: 0-3158), United States Marine Corps, for conspicuous gallantry and intrepidity while Commanding the First Marines, FIRST Marine Division (Reinforced), in action against enemy aggressor forces during the amphibious landing resulting in the capture of Inchon, Korea, on 15 September 1950 in the Inchon-Seoul Operation. His actions contributed materially to the success of this operation and were in keeping with the highest traditions of the Military Service. Namesakes and honors In addition to his military awards Puller has received numerous honors due to his Marine Corps service: On October 26, 2017, a groundbreaking ceremony was held for the Puller Veterans Care Center located in Vint Hill, Virginia. The facility was named in honor of Puller and his wife. The frigate was named after him. The headquarters building for 2nd Fleet Antiterrorism Security Team on Yorktown Naval Weapons Station in Yorktown, Virginia, is named Puller Hall in his honor. Route 33 in Middlesex County, Virginia, is named General Puller Highway. It is the county in which Puller is buried. On November 10, 2005, the United States Postal Service issued its Distinguished Marines stamps in which Puller was honored. The Marine Corps' mascot is perpetually named "Chesty Pullerton." (e.g. Chesty XIII). He is always a purebred English Bulldog. In 2012, Military Sealift Command announced that a Mobile Landing Platform will be named after Puller, . Death and legacy Following his retirement Puller lived in Saluda, Virginia, where he was later buried after his death on October 11, 1971, at Christ Church Parish Episcopal next to his wife. Puller remains a well-known figure in U.S. Marine Corps folklore, with both true and exaggerated tales of his experiences being constantly recounted among U.S. Marines. A common incantation in U.S. Marine Corps boot camp is to end one's day with the declaration, "Good night, Chesty, wherever you are!" Another common encouragement is "Chesty Puller never quit!" In U.S. Marine Corps recruit training and OCS cadences, Marines chant "It was good for Chesty Puller/And it's good enough for me" as well as "Tell Chesty Puller I did my best."—Chesty is symbolic of the esprit de corps of the Marines. Also, the recruits sing "Chesty Puller was a good Marine and a good Marine was </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> he." U.S. Marines, while doing push-ups, will tell each other to "do one for Chesty!" Puller insisted upon good equipment and discipline; once he came upon a second lieutenant who had ordered an enlisted man to salute him 100 times for missing a salute. Puller told the lieutenant, "You were absolutely correct in making him salute you 100 times, Lieutenant, but you know that an officer must return every salute he receives. Now return them all, and I will keep count." While on duty in Hawaii and inspecting the armory, Puller fined himself $100 for accidentally discharging a .45 caliber pistol indoors, although the charge for his men was only $20. In popular culture The book Marine!: The Life of Chesty Puller, , is about his life as a Marine. The book Chesty Puller's Rules of Success, , written by Bill Davis, Col, USMC (ret) explores 20 of Puller's "self-imposed principles of action" he gleaned from numerous meetings with the legendary General. The book Chesty: The Story of Lieutenant General Lewis B. Puller, USMC, , by Col. Jon T. Hoffman, USMCR, is a complete biography of his life, and winner of the Marine Corps Heritage Foundation's 2001 General Wallace M. Greene Award: Best Marine Corps History Book of the Year. John Ford directed Chesty: A Tribute to a Legend a 1976 film documentary. In the HBO miniseries The Pacific (2010), Puller is played by the American actor William Sadler. See also List of historically notable United States Marines References Bibliography External links Marine Corps History Division LtGen Lewis B. Puller Biography Devildog.com Recorded interview with Lieutenant General Lewis Burwell "Chesty" Puller Category:1898 births Category:1971 deaths Category:American Marine Corps personnel of World War II Category:American Marine Corps personnel of the Korean War Category:Military personnel from Virginia Category:People from West Point, Virginia Category:People of the Banana Wars Category:Recipients of the Air Medal Category:Recipients of the Distinguished Service Cross (United States) Category:Recipients of the Legion of Merit Category:Recipients of the Order of Military Merit (Korea) Category:Recipients of the Order of the Cloud and Banner Category:Recipients of the Navy Cross (United States) Category:Recipients of the Silver Star Category:United States Marine Corps generals </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> City Hall, Dordrecht The City Hall in Dordrecht is a renaissance building decorated in neoclassic style on the Stadhuisplein and over one of the city harbors; the Voorstraat haven. It is the seat of the city's government, which first received city rights in 1220, and today it is still the place where residents often hold their civic wedding ceremonies. History The town hall was built in the 14th century as a cloth hall, rebuilt in 1544 and given a fashionable neoclassic facade in 1635-43, in order to match the prestigious Amsterdam City Hall built in the same period. The 17th century look of the building was preserved during several restorations, but in 1835 a wooden tower was added to the top of the building. The lions at either side of the steps were made in 1841 by the Ironworks IJzergieterij L.J. Enthoven en Co, of The Hague. The right side of the building stretches over the Voorstraat haven with gothic arches dating from the 14th century, as is the cellar with its sculpted 14th century keystones in the vaulted ceilings. The clocks in the building were made by Pieter van Dormer en Sloterdijck in 1449 and by Gregorius Waghevens in 1514. The current use of the building is mostly for events and visitors. The municipal offices moved in 1975 to the Stadskantoor on the Spuiboulevard. References Category:Buildings and structures completed in 1544 Category:Towers completed in 1835 Dordrecht Category:Rijksmonuments in Dordrecht Category:1544 establishments in the Holy Roman Empire </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Samuel Yaw Samuel Yaw (born 2 February 1945) is a Ghanaian footballer. He competed in the men's tournament at the 1972 Summer Olympics. References Category:1945 births Category:Living people Category:Ghanaian footballers Category:Ghana international footballers Category:Olympic footballers of Ghana Category:Footballers at the 1972 Summer Olympics Category:Place of birth missing (living people) Category:Association football midfielders Category:Asante Kotoko SC players </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Hansborough, South Australia Hansborough is a locality along the former Morgan railway line adjacent to the Thiele Highway, in South Australia's Mid North region. It is situated 9 kilometres south-west of Eudunda and 18 kilometres north-east of Kapunda. The Light River runs through the locality. A town was surveyed in July 1865 and named after Frederick Hansborough Dutton (1812-1890), an early pastoralist and an overlander, who founded Anlaby Station, near Kapunda. It was declared as ceasing to exist on 13 August 1936. Boundaries were created for the part of the locality within the Light Regional Council on 16 March 2000 and for the part within the Regional Council of Goyder which includes the ceased Government Town of Hansborough on 24 August 2000. The Hundred of Neales School, later Freshwater Creek School, opened in 1927 in a former manager's residence on the Kingscourt property and closed in 1940. The 2016 Australian census which was conducted in August 2016 reports that Hansborough had a population of 42 people. See also David Moody (politician) References Category:Ghost towns in South Australia </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Roman Sadovsky Roman Sadovsky (born May 31, 1999) is a Canadian figure skater. He is the 2020 Canadian national champion, the 2019 NHK Trophy bronze medalist, and a three-time ISU Challenger Series medalist. He also won five medals on the ISU Junior Grand Prix series, including two gold medals. Personal life Sadovsky was born on May 31, 1999, in Toronto, Ontario, Canada. His parents moved to Canada from Ukraine. After attending Christ The King CES, he continued his education at Bill Crothers Secondary School. Career Early years Sadovsky began learning to skate at the age of five years in order to play hockey. Tracey Wainman started coaching him when he was eight. Another early coach was Allen Carson. 2012–2013 season Making his Junior Grand Prix debut, Sadovsky won a bronze medal in Lake Placid, New York, and placed tenth in Bled, Slovenia. He withdrew from the 2013 Canadian Championships due to a metatarsal fracture in his right foot on a growth plate near the toe. 2013–2014 season Coached by Tracey Wainman and Grzegorz Filipowski at the York Region Skating Academy in Richmond Hill, Ontario, Sadovsky competed in two events of the 2013 Junior Grand Prix series, placing fourteenth in Riga, Latvia, and eighth in Minsk, Belarus. Nationally, he appeared on the senior level, finishing eighth at the 2014 Canadian Championships. He was selected for the 2014 World Junior Championships in Sofia, Bulgaria. Ranked fourteenth in the short program and twelfth in the free skate, he finished thirteenth overall. 2014–2015 season Sadovsky's first assignment of the 2014 Junior Grand Prix series was in Ostrava, Czech Republic. Ranked first in the short program and third in the free skate, he finished first overall by a margin of 3.39 points over the silver medalist, Alexander Samarin. After the event, Sadovsky said his goal was to develop a solid triple Axel. He then went on to place fourth at his second JGP event in Dresden, Germany. With those results, Sadovsky qualified for the 2014 JGP Final and placed fifth. At the 2015 Canadian Championships, Sadovsky placed fourth. He ended his season with a fourteenth-place finish at the 2015 World Junior Championships. 2015–2016 season In the 2015 Junior Grand Prix season, Sadovsky was assigned to the first event, held in Bratislava, Slovakia. With a quad Salchow in the free skate — Sadovsky's first quad in competition — he won the gold medal with a total score 2.87 points ahead of Vincent Zhou of the United States. He then went on to win bronze at his second JGP event, in Toruń, Poland. These results qualified him for the 2015 JGP Final, where he was sixth. He represented Canada at the 2016 Winter Youth Olympics and finished fourth. He was coached by Wainman and Filipowski. 2016–2017 season Sadovsky changed coaches, joining Brian Orser and Lee Barkell at the Toronto Cricket, Skating and Curling Club. He placed ninth at the 2017 Canadian Championships. Ranked ninth in the short program and twenty-third in the free skate, he finished seventeenth at the 2017 World Junior Championships in Taipei, Taiwan. 2017–2018 season Sadovsky switched back to Wainman and Filipowski. Making his Grand Prix debut, he placed tenth at the 2017 Skate America. He finished seventh at the 2018 Canadian Championships. 2018–2019 season Sadovsky began the season at the 2018 CS Autumn Classic International, where he placed fourth in both the short and free programs, winning the bronze medal, his first senior medal. Sadovsky landed his first quad toe loop jump in competition. He then placed twelfth at the 2018 Skate Canada International, after struggling in both of his programs, and won </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> the silver medal at the 2018 Inge Solar Memorial. Sadovsky placed fifth in the short program at the 2019 Canadian Championships, executing his open quadruple Salchow-triple toe loop combination cleanly, but popping a planned quad toe loop and receiving a negative Grade of Execution on his triple Axel. He dropped to seventh place overall following a difficult free skate. 2019–2020 season Starting out the season on the Challenger series, Sadovsky won the bronze medal at the 2019 CS Finlandia Trophy after placing third in the short program and fourth in the free skate. He called the short program "the best I've ever done", but said "the long could have been better." Sadovsky placed tenth at Skate Canada International, his first Grand Prix assignment for the season. Competing next at the 2019 NHK Trophy, Sadovsky placed fourth in the short program despite making two jump errors, one of them costing him the second part of his combination. He was second in the free skate, making only two minor errors on his second triple Axel and closing triple Lutz, and placed third overall, taking his first even Grand Prix medal. He remarked "I’m still new to the senior Grand Prix circuit and there are some mistakes that need to be fixed. I hope this success on the senior Grand Prix will not be my last." Entering the 2020 Canadian Championships as a contender for the title, Sadovsky placed third in the short program behind Keegan Messing and Nam Nguyen, having had two of his jumps called as underrotated. He then won the free skate, with only two minor jump errors on his triple loop and second triple Axel, while Messing and Nguyen both struggled. Sadovsky won his first national title by a margin of over seventeen points. Sadovsky was named to one of Canada's three men's berths at the 2020 Four Continents Championships, but Skate Canada declined to immediately decide who would be Canada's sole men's representative at the 2020 World Championships in Montreal. Sadovsky placed a poor sixteenth at Four Continents, and thus was not chosen for the World Championships, instead claimed by Nguyen. Programs Competitive highlights GP: Grand Prix; CS: Challenger Series; JGP: Junior Grand Prix Detailed results Small medals for short and free programs awarded only at ISU Championships. At team events, medals awarded for team results only. References External links Category:1999 births Category:Canadian male single skaters Category:Canadian people of Ukrainian descent Category:Competitors at the 2016 Winter Youth Olympics Category:Living people Category:Sportspeople from Toronto </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Glaphyria oriola Glaphyria oriola is a moth in the family Crambidae. It was described by Harrison Gray Dyar Jr. in 1914. It is found in Panama. References Category:Moths described in 1914 Category:Glaphyriini </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Dryden Water Aerodrome Dryden Water Aerodrome, , is located south of Dryden, Ontario, Canada. See also Dryden Regional Airport References Category:Registered aerodromes in Kenora District Category:Transport in Dryden, Ontario Category:Seaplane bases in Ontario </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Frederico Tadewald Frederico Tadewald (born 19 January 1908, date of death unknown) was a Brazilian rower. He competed in the men's eight event at the 1936 Summer Olympics. References Category:1908 births Category:Year of death missing Category:Brazilian male rowers Category:Olympic rowers of Brazil Category:Rowers at the 1936 Summer Olympics Category:People from Rio Grande do Sul </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Columbus Post Office The Columbus Post Office is the main post office in Columbus, Wisconsin. The post office was built in 1938 by the Public Works Administration and opened in 1939. The brick building was designed in the Art Moderne style. Arnold Blanch painted a mural in the post office in 1940 to honor the founding of Columbus; the mural includes the city's first log cabin settlement, farmers and grains from the area, and depictions of typical 1930s residents of the city. The post office was added to the National Register of Historic Places on October 24, 2000. References Category:Post office buildings on the National Register of Historic Places in Wisconsin Category:Government buildings completed in 1938 Category:Buildings and structures in Columbia County, Wisconsin Category:Streamline Moderne architecture in Wisconsin Category:Columbus, Wisconsin Category:1938 establishments in Wisconsin Category:National Register of Historic Places in Columbia County, Wisconsin </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> The Harvest Shall Come The Harvest Shall Come is a 1942 British documentary film about agricultural work between 1900 and World War II, using the story of a farm laborer to illustrate the importance of agriculture, and the importance of supporting workers in this occupation. The film, produced by Basil Wright and directed by Max Anderson, was well received. Production The Harvest Shall Come was entirely filmed on location, including the interior scenes. The film was mostly shot at Hill House Farm, Needham Market in Suffolk. Other scenes were filmed at Badley Hall and in Willisham. Many of the extras were local people. The 33-minute documentary, with music by William Alwyn, was sponsored by the Ministry of Information. The Harvest Shall Come was made for Imperial Chemical Industries, and makes the case for fertilizing the soil to improve productivity. Synopsis The opening titles appear over scenes of farm workers stacking hay. The film opens in 1900 when Tom Grimwood as a boy leaves his family cottage carrying his trunk to take a job on a farm for a weekly wage of 2/6 plus keep. The commentary notes that farm workers receive the lowest pay and worst housing of all workers, but that there had been improvements since 1870. Tom is taught to work the plough, the harrow and other horse-drawn machinery, and learns skills such as milking and scything. At the age of 21 Tom gets a new job, which pays 14/- a week and comes with a run-down cottage, so he can afford to marry his sweetheart. The film moves forward to a scene where the couple have three small children, but still have no sink in the cottage. In World War I (1914–18) farm workers manage to gain better wages. After the war, despite the promises of politicians, wages sink to just 30/- per week. Although Tom wants to leave for a better job, he cannot afford to move from the cottage. The film shows that in the interwar period standards of living in the country generally improved, but agricultural wages declined when imported corn depressed prices. Tom is laid off for a while and forced to dig ditches "on relief". His son decides to move to the town, where there are better opportunities. With the outbreak of World War II in 1939 imports are restricted and as much land as possible must be brought under cultivation. Agricultural labor conditions improve again and wages rise to 60/- a week, although prices also rise. A radio announcer says that the government will look after farm workers. Tom's wife is skeptical, saying "They said all that in the last war". The film ends with scenes illustrating the importance of agriculture around the world. Reception The Spectator said the film was "an excellent piece of work, notable for its effective and truthful presentation of the vital subject of life in this country on the land." The Harvest Shall Come was named "Documentary of the Month" by Documentary News Letter, which called it "the first genuine story film made with the documentary purpose and by documentary method." It praised Max Anderson as one of the best documentary makers to appear for many years and called John Slater "an outstanding interpreter of working class character." Sir Arthur Elton of the films division of the Ministry of Information said in 1943 that the film was one of the best documentary films that had been produced. The Harvest Shall Come has been called "a far-sighted educational treatment of a major social problem." Cast John Slater: Tom Grimwood Eileen Beldon: Lil Victor Woolf: Bill Ernest Borrow: </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Squire Richard George: Farmer Edmund Willard: Commentator Bruce Belfrage: Commentator See also The Battle of the Harvests, a 1942 Canadian documentary References Citations Sources External links Category:1942 films Category:1940s documentary films Category:British films Category:English-language films Category:Black-and-white documentary films Category:Documentary films about agriculture Category:British documentary films Category:History of agriculture in the United Kingdom Category:Films directed by Max Anderson Category:Films scored by William Alwyn Category:British black-and-white films </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Febel Febel is the surname of the following people: Fritz Febel (1910–1969), German-American football player and coach Reinhard Febel (born 1952), German composer Category:Surnames of German origin </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> P. monstrosus P. monstrosus may refer to: Paraharmochirus monstrosus, a jumping spider species Poltys monstrosus, an orb-weaverspider species in the genus Poltys See also Monstrosus (disambiguation) </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Serrano v. Priest Serrano v. Priest refers to three cases regarding the financing of public schools in California that were decided by the California Supreme Court: Serrano v. Priest, (1971) (Serrano I); Serrano v. Priest, (1976) (Serrano II); and Serrano v. Priest, (1977) (Serrano III). The Serrano cases Serrano I (1971) Initiated in 1968 in the Superior Court of Los Angeles County, Serrano v. Priest (John Serrano was a parent of one of several Los Angeles public school students; Ivy Baker Priest was the California State Treasurer at the time) set forth three causes of action (quotes from the decision). "[As] a direct result of the financing scheme they are required to pay a higher tax rate than [taxpayers] in many other school districts in order to obtain for their children the same or lesser educational opportunities afforded children in those other districts." "[That] an actual controversy has arisen and now exists between the parties as to the validity and constitutionality of the financing scheme under the Fourteenth Amendment of the United States Constitution and under the California Constitution." In an opinion by Justice Raymond L. Sullivan, the Court agreed with the plaintiffs, largely on equal-protection grounds, and returned the case to the trial court for further proceedings. As Sullivan summarizes, "We are called upon to determine whether the California public school financing system, with its substantial dependence on local property taxes and resultant wide disparities in school revenue, violates the equal protection clause of the Fourteenth Amendment. We have determined that this funding scheme invidiously discriminates against the poor because it makes the quality of a child's education a function of the wealth of his parents and neighbors. Recognizing as we must that the right to an education in our public schools is a fundamental interest which cannot be conditioned on wealth, we can discern no compelling state purpose necessitating the present method of financing. We have concluded, therefore, that such a system cannot withstand constitutional challenge and must fall before the equal protection clause." Serrano II (1976) In San Antonio Independent School District v. Rodriguez (1973), the Supreme Court of the United States reversed a similar decision by a Texas District Court, which like Serrano I had been decided on Fourteenth Amendment equal-protection grounds. In Serrano I, however, the California Supreme Court had relied in addition on California's constitution, and in Serrano II they affirmed that basis, protecting the Serrano decisions from Rodriguez. The Serrano II decision also held that the legislative response to Serrano I was insufficient, and affirmed the trial court's order requiring that wealth-based funding disparities between district be reduced to less than $100 by 1980. Serrano III (1977) Serrano III dealt primarily with attorneys' fees, but in passing affirmed the trial court's response to the Serrano II decision, including a six-year timetable for bringing the funding system into compliance. Proposition 13 The California State Legislature's response to Serrano I and Serrano II was significantly constrained by the passage of Proposition 13 in 1978, which reduced property-tax revenues and imposed a 2/3-majority vote requirement for statewide tax increases. The initial property-tax-based solution was replaced by a funding scheme that relied more heavily on state (as opposed to district) revenue, which has remained in effect, with occasional adjustments, ever since. Compliance (1983) In 1983, the Los Angeles County Superior Court found, on remand, that the requirements of Serrano II had been sufficiently met, allowing a relatively small number of residual districts to retain a higher level of funding, based on well-above-average local property taxes. See also San Antonio Independent School District v. Rodriguez (Texas) Edgewood Independent </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> School District v. Kirby (Texas) Abbott v. Burke (New Jersey) References Further reading Dollars and Sense: A Simple Approach to School Finance, California Little Hoover Commission, 1997 Hanushek, Eric A., and Alfred A. Lindseth. 2009. Schoolhouses, courthouses, and statehouses: Solving the funding-achievement puzzle in America's public schools. Princeton, NJ: Princeton University Press External links Text of Serrano I decision via FindLaw Text of Serrano II decision via FindLaw Text of Serrano III decision via FindLaw Category:Public education in California Category:Education finance in the United States Category:Economic inequality in the United States Category:Taxation and redistribution Category:California state case law Category:United States education case law Category:1971 in United States case law Category:1976 in United States case law Category:Los Angeles Unified School District Category:Law articles needing an infobox </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Metsaküla, Põhja-Pärnumaa Parish Metsaküla is a village in Põhja-Pärnumaa Parish, Pärnu County in western-central Estonia. Category:Villages in Pärnu County </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> List of tourist attractions in Hamilton, Ontario Hamilton, Ontario has a large variety of historical sites and cultural and educational institutions. Historical sites and museums 31 Service Battalion Museum, accredited military museum, Heritage Museum; dedicated to documenting Combat Service Support in Hamilton; largest collection of the Canadian Women's Army Corps artifacts; located next to HMCS Haida Auchmar, historic estate of the Honourable Isaac Buchanan Battlefield House Museum, Stoney Creek Canadian Warplane Heritage Museum, home of one of the last two remaining operational Lancaster bombers; also in operation Spitfire, Hawker Hurricane, Bristol Bolingbroke Dundurn Castle, including the Hamilton Military Museum and Dundurn Park, west end; home of Sir Allan Napier MacNab, former Prime Minister of Upper Canada Erland Lee Museum, birthplace of Women's Institutes, Upper Stoney Creek Hamilton Children's Museum, east end Hamilton Farmer's Market, founded in 1837 Hamilton Museum of Steam and Technology, east end HMCS Haida, National Historic Site, historic naval ship; Canada's most famous warship and the last remaining Tribal Class in the world Nash-Jackson House, at Stoney Creek Battlefield Park Ottawa Street North, textile district, voted one of Canada's favourite streets in the 2011 'Great Places in Canada' contest; ranked in Top 5 spots for antique shopping in Canada by CAA Magazine Royal Hamilton Light Infantry Heritage Museum, downtown Whitehern Historic House & Garden, downtown Cultural institutions Art Cotton Factory Arctic Experience McNaught Gallery Art Gallery of Hamilton, downtown; second largest permanent collection in Ontario, and third largest in Canada James Street North Art District McMaster Museum of Art, west end Music Bach Elgar Choir Hamilton Philharmonic Orchestra Hess Village, commercial and entertainment hub in historic buildings Theatre Hamilton Theatre Inc., musical theatre Other popular attractions Festivals Festival of Friends, founded in 1975, the largest annual free music event in Canada Sports Around the Bay Road Race, the longest continuously held long distance foot race in North America (since 1894) Canadian Football Hall of Fame, downtown CANUSA Games, Hamilton is twinned with Flint, Michigan, and its amateur athletes have competed in these games since 1958 Jukasa Motor Speedway, a 5/8-mile oval auto racing track, Cayuga, Ontario Flamboro Downs and Flamboro Slots, horse racing as well as car racing Parks, trails and waterfront African Lion Safari, Flamborough Bayfront Park, Pier 4 Park, Harbour West, Cootes Paradise, waterfront trail Beach strip and lighthouse pier, beach trail Bruce Trail, Stoney Creek, Hamilton, Dundas, Flamborough Dundas Valley Conservation Area, Dundas Gage Park, large historical park in the middle of the city Hamilton Farmer's Market Pier 4 Park Royal Botanical Gardens, west end Westfield Heritage Centre, Flamborough References * Hamilton, Ontario </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Bayanjargalan Bayanjargalan is the name of two sums (districts) in Mongolia: Bayanjargalan, Dundgovi Bayanjargalan, Töv </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Nectopanope Nectopanope rhodobaphes is a species of crabs in the family Xanthidae, the only species in the genus Nectopanope. References Category:Xanthoidea Category:Monotypic arthropod genera </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Ravinia Festival The Ravinia Festival is the oldest outdoor music festival in the United States, with a series of outdoor concerts and performances held every summer from June to September. In Ravinia Park's first summer of 1905, it hosted the New York Philharmonic, and the prairie style Martin Theater dates from this time period. It has been the summer home of the Chicago Symphony Orchestra (CSO) since 1936. Located in Highland Park, Illinois, the festival operates on the grounds of the Ravinia Park, with a variety of outdoor and indoor performing arts facilities. The Ravinia neighborhood, once an incorporated village before annexation in 1899, is known as Ravinia, and retained its own post office until autumn 2010. The business district on Roger Williams Ave., within walking distance from the Ravinia Festival grounds, includes neighborhood service businesses and restaurants. Ravinia takes its name from the ravines found nearby along the shoreline of Lake Michigan. Performance and other venues The Pavilion, a 3,400-seat venue where the park's major music events and concerts, including Chicago Symphony Orchestra performances, are held. The Martin Theatre, an 850-seat indoor hall often used for chamber music, semi-staged opera performances, Martinis at the Martin cabaret series, and other shows. Bennett Gordon Hall, the 450-seat home of Ravinia's Steans Music Institute, the $10 BGH Classics Series, and also used for pre-concert discussions and preview concerts. Ravinia's Steans Music Institute is the Ravinia Festival's pre-professional summer conservatory program. Three programs comprise the Institute's summer season: the program for jazz; the program for piano and strings, and the program for singers. Overview In 1904, the A.C. Frost Company created Ravinia as an amusement park intended to lure riders to the fledgling Chicago and Milwaukee Electric Railroad. The amusement park boasted a baseball diamond, electric fountain and refectory or casino building with dining rooms and a dance floor. The prairie-style Martin Theatre (then called Ravinia Theatre) is the only building on the grounds that dates back to that original construction. When the park's existence became jeopardized following the railroad's bankruptcy, local residents (for the most part Chicago businessmen) formed a corporation in 1911 to purchase and operate the park. Music was a confirmed summer activity from then on, except for a brief hiatus during the Depression. Over 100 years later, Ravinia Festival is the oldest outdoor music festival in North America and is lauded for presenting world-class music. The festival attracts about 600,000 listeners to some 120 to 150 events that span all genres from classical music to jazz to music theater over each three-month summer season. Over the years, the festival has hosted many famous artists. In addition to symphony concerts, often with guest soloists, the festival presents opera, jazz, blues, folk, rock, and popular music performances, plus ballet, drama, and educational programs which take place year-round. Recent artists who have performed at Ravinia include John Legend, Aretha Franklin, Bryan Ferry, Diana Ross, Maroon 5, Common, Carrie Underwood, Tony Bennett, Lady Gaga, Josh Groban, Dolly Parton, Sheryl Crow, Patti LaBelle, Andrew Bird, Darius Rucker, Mary J. Blige, Gladys Knight, Santana, Stevie Nicks, Patti LuPone, Smokey Robinson, and John Mellencamp. Grounds For most attendees, Ravinia is experienced on the 36 acre (150,000 m²) parkland and lawn. Ravinia is one of the few concert venues in the country to allow full meals to be brought in and consumed at concerts, even allowing alcoholic beverages. Accordingly, most grocery stores and specialty restaurants in and around the Highland Park area offer ready-to-eat "Ravinia picnics" for purchase. The park is served by the Metra commuter railroad station Ravinia Park outside the front gate with special stops </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> before and after concerts. (The noted British conductor Sir Thomas Beecham, who guest-conducted the CSO there in 1940, referred to Ravinia as "the only railway station with a resident orchestra.") Visitors get dropped off and picked up right at the front gate. Attendance often tops 600,000 annually. Artistic leadership Walter Hendl, artistic director (1959-63) Seiji Ozawa, Music Director (1964-68)and principal conductor (1969) István Kertész,principal conductor (1970-72) James Levine, Music Director (1973-93) Christoph Eschenbach, Music Director (1995-2003) James Conlon, Music Director (2005-15) Ramsey Lewis, Artistic Director, Jazz at Ravinia Welz Kauffman, President & CEO (2000–present) Marin Alsop, Artistic Curator (2018-2019) and Chief Conductor and Curator (2020-present) James Levine was named "Conductor Laureate" in April 2017, to begin performances in summer 2018. On December 4, 2017, the Ravinia Festival severed all ties with Levine, in the wake of sexual abuse allegations against him, dating back to decades earlier at the Ravinia Festival. See also List of contemporary amphitheatres List of opera festivals References External links Chicago Sun-Times' article on the Festival's Centennial Ravinia Festival Association Records at Newberry Library Category:Amphitheaters in the United States Category:Amusement parks in Illinois Category:Music venues in Illinois Category:Music festivals in Illinois Category:Opera festivals Category:Highland Park, Illinois Category:Tourist attractions in Lake County, Illinois Category:Buildings and structures in Lake County, Illinois Category:Music festivals established in 1904 Category:1904 establishments in Illinois </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Robert Courtney Robert Ray Courtney (born September 15, 1952) is an American former pharmacist from Kansas City, Missouri. In 2002, he pleaded guilty to intentionally diluting several doses of chemotherapy drugs and was sentenced to 30 years in federal prison. He is currently serving his sentence at Big Spring Federal Correctional Institution in Big Spring, Texas. Early life and education Courtney was born in Hays, Kansas. His father was a traveling minister based in Scott City, and described Courtney as an "ideal son." He played the trombone at Wichita South High School. Courtney graduated from the School of Pharmacy at University of Missouri–Kansas City in 1975. Adult life In 1986, Courtney became the owner of Research Medical Tower Pharmacy in Kansas City, where he had worked for some time. He primarily mixed intravenous drugs. Before his arrest, Courtney served as a deacon at Northland Cathedral, an Assemblies of God megachurch in Kansas City. In 1992, he and his first wife divorced; Courtney retained custody of their two daughters. His second marriage lasted four or five days and was later annulled. In 1994 his third wife, Laura Courtney, gave birth to twins. In August 2001, the same month he was arrested, Courtney held total assets worth $18.7 million. Dilution fraud In 1990, Courtney began purchasing pharmaceuticals on the gray market and using them to fill prescriptions at his pharmacy. In time he began diluting prescriptions to increase profits. Both practices were illegal. In 1998, Darryl Ashley, an Eli Lilly sales representative noticed Courtney was selling three times the amount of the cancer drug Gemzar than he'd bought. Lilly initiated an internal investigation but found no evidence of illegality and closed the investigation without further action. In early 2001, Ashley once again voiced his concerns to a nurse who worked for Dr. Verda Hunter (now Hicks), an oncologist in Courtney's building, who was also one of Courtney's customers. Hunter noticed that many of her patients were only suffering mild side effects, and their condition didn't seem to be improving. Hunter had medication that had been supplied by Courtney tested. That test showed that the sample contained less than one-third of the drug prescribed, and upon receiving the test results back, Hunter immediately notified the FBI. Federal investigators initially didn't believe that a pharmacist would do something so egregious. However, when Hunter showed them the test results, they realized that she was telling the truth. Hunter submitted seven additional samples for testing by the FDA's forensic chemistry lab. Tests on those samples revealed that they contained as little as 15 percent of the prescribed dosage, and at most only half of it. They immediately knew that they had to move quickly. While health care fraud cases normally take years to build, the investigators knew they didn't have that long. Investigators believed that Courtney took a base dose of chemotherapy drugs and split it between three prescriptions, then sold them to oncologists for the same price as a full dose. He took advantage of the fact that oncologists are usually concerned mainly with chemotherapy's effects on the body, not the amount of the dose. Investigators persuaded Hunter to help them in a sting operation. Hunter gave Courtney several prescriptions for fictitious patients. After Courtney mixed the drugs and sent them to Hunter's office, federal agents had them tested. The samples contained less than half of the prescribed dosage, and in some cases contained less than one percent of the active drug. On August 13, 2001, federal agents raided Research Medical Tower Pharmacy. A day later, Courtney surrendered to authorities and was charged with one </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> count of adulterating and misbranding medication. Investigators reported that before turning himself in, Courtney gave $80,000 in cash, and more than 100 doses of Prozac to his wife. Faced with the evidence, Courtney gave investigators a list of three medications that he diluted, and a list of 34 affected patients. He claimed to have only started diluting drugs a few months ago, a claim no one believed. He openly admitted he did it to pay off a $1 million donation to his church's building fund. On August 23, 2001, Courtney was indicted on 20 counts of tampering with consumer products and adulterating and misbranding drugs. Many patients and survivors wanted him charged with murder, but federal prosecutors believed a murder charge would be hard to prove since many patients were suffering from late-stage cancer. Courtney also was named as defendant in approximately 300 suits for fraud and wrongful death. In one case a jury awarded the plaintiff, Georgia Hayes, a judgment in the amount of $2.2 billion. Although Hayes knew she would likely never see that money because his assets had been frozen, she wanted to send a message that this type of deceit was not worth the cost. Eli Lilly and Bristol Myers-Squibb were named in several of the civil suits. Eli Lilly ultimately settled the suits for $48 million, while Bristol Myers-Squibb paid $24 million. Facing the prospect of life in prison if convicted at trial, on February 20, 2002, Courtney pleaded guilty to 20 federal counts of tampering and adulterating the chemotherapy drugs Taxol and Gemzar. He also acknowledged that he and his corporation, Courtney Pharmacy Inc., had weakened drugs, conspired to traffic in stolen drugs and caused the filing of false Medicare claims. According to law enforcement estimates, as well as his own confession, from 1992 to 2001 Courtney diluted 98,000 prescriptions from 400 doctors, which were given to 4,200 patients. Courtney admitted to diluting 72 different kinds of drugs. Besides chemotherapy treatments, he admitted diluting medications for diabetes and AIDS patients, as well as fertility treatments. On December 5, 2002, Courtney was sentenced to 30 years in federal prison. In 2008, an episode of American Greed, entitled "Deadly ℞ For Greed", recounted Courtney's crimes, trial and conviction. Courtney, Federal Bureau of Prisons inmate number 14536-045, is currently serving his sentence at Big Spring Federal Correctional Institutuon, after starting his sentence at Gilmer Federal Correctional Institution near Glenville, West Virginia. His earliest possible release date is November 20, 2027, when he will be 75 years old. References See also Counterfeit medications Category:American fraudsters Category:Eli Lilly and Company people Category:American pharmacists Category:People from Hays, Kansas Category:1952 births Category:Living people Category:University of Missouri–Kansas City alumni Category:American people convicted of fraud </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Strophostyles leiosperma Strophostyles leiosperma, known as slickseed fuzzybean, or smoothseed / small-flower wildbean is a species of herbaceous, vining legume native to the central to western U.S. It occurs west to Colorado and New Mexico, east to Louisiana, south to Mexico, and north to Minnesota. It is most easily distinguished from the other two Strophostyles species by the abundance of small silky hairs on its leaves and pods, and small pea-shaped flowers with a much reduced keel that is largely hidden by the wing petals. This species is an annual to short-lived perennial. All parts tend to be smaller for S. leiosperma in general than its congeners, and it is a more diminutive plant overall. The leaflets are typically thin and rarely lobed (never deeply lobed). Unlike its congeners, its seeds rarely have a waxy, hairy covering, and it tends to occur in drier sites. Likewise, the specific epithet leiosperma means "smooth seed." It is also the most likely of these species to be capable of self-fertilization. References Category:Phaseoleae Category:Legumes </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Marcia Joanne Bennett Marcia Joanne Bennett (born June 9, 1945, sometimes credited as M.J. Bennett), is an American writer of fantasy and science fiction novels, including the Ni-Lach series and the novel Yaril's Children. Biography Born in Norwich, NY, daughter of Richard (a carpenter) and Reatha (née Albright). After graduating from Albany Business College in 1965, she worked in the banking industry as first a secretary and then a teller. She opened a craft shop in Earlville, New York in 1972, which came to serve as her writing studio. About her work, Bennett states, "For me writing was a natural progression from years of being an avid reader. That, coupled with a penchant for daydreaming, led me to a hobby that quickly became an addiction. I choose to write fantasy and science fiction because they give me a freedom I do not find in other types of literature." Themes and critical response In an interview with Contemporary Authors Online, Bennett said of her work, "Friendship and tolerance are my main themes." Though her novels produced lukewarm reactions from some critics (The Encyclopedia of Science Fiction noting, "The local-colour quotient is high, though the sequence [of the Ni-Lach series] itself is unremarkable" ), other reviewers had more positive reactions. Of her debut novel, Where the Ni-Lach, Publishers Weekly wrote, "Bennett's well-depicted characters, sustained sense of mystery and several clever plot twists make for a compelling read." In a review of the second book in the series, The Emergency Librarian said, "The themes of understanding differences, personal responsibility and loyalty to friends, and a group of engaging protagonists enrich the standard adventure plot." The third novel in the series, Beyond the Draak's Teeth received a "recommended" review from Library Journal: "The values of freedom and friendship hang in a delicate balance in this sf/fantasy adventure." Bibliography Where the Ni-Lach (1983) Shadow Singer (1984) Beyond the Draak's Teeth (1986) Yaril's Children (1988) Seeking the Dream Brother (1989) References External links at the Internet Science Fiction Database at the Science Fiction and Fantasy Book Review Index at The Encyclopedia of Science Fiction Category:American science fiction writers Category:Living people Category:American women novelists Category:American fantasy writers Category:20th-century American novelists Category:1945 births Category:Women science fiction and fantasy writers Category:20th-century American women writers Category:People from Norwich, New York Category:Novelists from New York (state) </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Flute sonata A flute sonata is a sonata usually for flute and piano, though occasionally other accompanying instruments may be used. Flute sonatas in the Baroque period were very often accompanied in the form of basso continuo. List of flute sonatas George Antheil Sonata for flute and piano (1951) Malcolm Arnold Sonata for flute and piano, Op.121 (1977) Carl Philipp Emanuel Bach Sonata in A minor for solo flute (H. (Helm) 562/Wq. (Wotquenne) 132) (1747) 14 sonatas for flute and continuo Sonata in G minor for flute and harpsichord, BWV 1020 Johann Christian Bach Sonatas for keyboard with flute or violin, Op. 16 Johann Sebastian Bach Sonata in B minor for flute and harpsichord, BWV 1030 Sonata in E-flat major for flute and harpsichord, BWV 1031 Sonata in A major for flute and harpsichord, BWV 1032 Sonata in C major for flute and basso continuo, BWV 1033 Sonata in E minor for flute and basso continuo, BWV 1034 Sonata in E major for flute and basso continuo, BWV 1035 Partita in A minor for solo flute, BWV 1013 Arnold Bax Sonata for flute and harp (1928) Lennox Berkeley Sonata for flute and piano (1978) Ludwig van Beethoven (attributed) Flute Sonata in B-flat major, Anh 4 Flint Juventino Beppe Flute Sonata No.1 Op.40 (1998) Pierre Boulez Sonatina for flute and piano (1946) (1 serial movement) York Bowen Flute Sonata, Op. 120 (1946) Robert J. Bradshaw Sonata No. 2 In My Collection for flute and piano Edison Denisov Sonata for flute and piano (1960) Sonata for flute and guitar (1977) Sonata for flute solo (1982) Sonata for flute and harp (1983) Pierre Max Dubois Sonata for flute and piano Henri Dutilleux Sonatina for flute and piano (1943) Jindřich Feld Sonata for flute and piano (1957) César Franck Sonata for Violin and Piano in A major (transcribed for flute) (1886) Glenn Gould Sonata for Flute and Piano (1950). This work is an arrangement of Gould's Sonata for Bassoon and Piano. Jorge Grundman Warhol in Springtime. Sonata for Flute and Piano (2011) George Frideric Handel Sonata for flute in E minor, Op. 1, No.1A Sonata for flute in E minor, Op. 1, No.1B Sonata for Flute in G major Op. 1, No.5 Sonata for Flute in B minor Op. 1, No.9 Hans Werner Henze Sonatina for flute and piano (1947) Paul Hindemith Sonata for flute and piano (1936) Bertold Hummel Sonatina for flute and piano, Op. 107a (2001) Johann Nepomuk Hummel Sonata in D, Op. 50 (c1810–14) Sonata in A, Op. 64 (c1814–15) Philipp Jarnach Sonatina for flute and piano, Op. 12 (1919) Sándor Jemnitz Sonata for flute and piano, Op. 27 (1930–31) Paul Juon Sonata for flute and piano in F, Op. 78 (1924) Sigfrid Karg-Elert Sonata for flute and piano in B flat, Op. 121 (1918) Sonata Appassionata for flute solo in F sharp, Op.140 (1917) Charles Koechlin Sonata for flute and piano, Op.52 (1911-13) Sonata for 2 flutes, Op.75 (1918-20) Three sonatines for solo flute, Op.184 (1942) Jean-Marie Leclair Sonata for flute and harpsichord No.1 in B major, Book 1, No.2 Sonata for flute and harpsichord No.2 in E minor, Book 1, No.6 Sonata for flute and harpsichord No.3 in E minor, Book 2, No. 1 Sonata for flute and harpsichord No.4 in C major, Book 2, No. 3 Sonata for flute and harpsichord No.5 in G major, Book 2, No. 5 Sonata for flute and harpsichord No.6 in B minor, Book 2, No. 11 Sonata for flute and harpsichord No.7 in E minor, Book 4, No. 2 Sonata for flute and harpsichord No.8 in G major, Book 4, No. 7 </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Dieter Lehnhoff Sonata Porteña for flute and piano, Op. 35 (2013) Bohuslav Martinů Sonata for flute and piano, Halbreich 306 (1945) Peter Mieg Sonata for flute and piano (1963) Darius Milhaud Sonatina for flute and piano, Op. 76 (1922) Ignaz Moscheles Sonata for flute and piano in A, Op. 44 (1819) Sonata for flute and piano in G, Op. 79 (1828) Wolfgang Amadeus Mozart Sonata in B-flat for Keyboard and Violin or Flute, K. 10 Sonata in G for Keyboard and Violin or Flute, K. 11 Sonata in A for Keyboard and Violin or Flute, K. 12 Sonata in F for Keyboard and Violin or Flute, K. 13 Sonata in C for Keyboard and Violin or Flute, K. 14 Sonata in B-flat for Keyboard and Violin or Flute, K. 15 Jules Mouquet La Flûte de Pan, Sonata, Op. 15 Robert Muczynski Sonata for flute and piano, Op. 14 (1961) Gabriel Pierné Sonata for flute and piano, Op. 36 (1900) Willem Pijper Sonata for flute and piano (1925) Walter Piston Sonata for flute and piano (1930) Francis Poulenc Flute sonata, Schmidt 164 (1956–7) Sergei Prokofiev Flute sonata in D, Op. 94 (1943) Einojuhani Rautavaara Sonata for flute and guitar (1975) Carl Reinecke Undine, Flute sonata, Op. 167 (1882) Ferdinand Ries Flute Sonata in G major, Op. 48 (pub. 1815) Divertimento for Piano and Flute in G major, Op.62 (1815, pub. 1819) Flute Sonata in C major, Op. 76, No. 1 (1816, pub. ca 1817/18) Flute Sonata in G major, Op. 76, No. 2 (1817, pub. ca 1817/18) Flute Sonata in G major, Op. 87 (pub. 1819) Flute Sonata in E-flat major, Op. 169 Sonate sentimentale (1814, pub. 1834) R. Murray Schafer Sonatina for flute and harpsichord (or piano) (1976) Erwin Schulhoff Sonata for flute and piano (1927) Leo Smit Sonata for flute and piano (1939-43) Otar Taktakishvili Sonata for flute and piano (1968) Hiroki Yamada Flute Sonata (2010-12) See also Flute concerto Flute repertory References </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Henry Coalter Cabell House Henry Coalter Cabell House is a historic home located in Richmond, Virginia. It was built in 1847, and was originally built as a two-story, "L"-shaped Greek Revival style brick dwelling. It was subsequently added to throughout the 19th century. It consists of a large central section with two-story portico, flanked by two smaller wings. The house features a portico supported by four columns with "Egyptian" lotus flower capitals. It was the home of Henry Coalter Cabell, son of Governor William H. Cabell, and a Confederate veteran and leading member of the Richmond Bar. It was listed on the National Register of Historic Places in 1972. References External links 116 South Third Street (House), Richmond, Independent City, VA: 1 photo at Historic American Buildings Survey Category:Historic American Buildings Survey in Virginia Category:Houses on the National Register of Historic Places in Virginia Category:Greek Revival houses in Virginia Category:Houses completed in 1847 Category:Houses in Richmond, Virginia Category:National Register of Historic Places in Richmond, Virginia Category:1847 establishments in Virginia </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Marseille Metro The Marseille Metro () is a rapid transit system serving Marseille, in Southern France. As of 2020, the system comprises two lines, partly underground, serving 29 stations, with an overall route length of . Line 1 opened in 1977, followed by Line 2 in 1984. Two stations, Saint-Charles and Castellane, provide interchange between each lines. The system's MPM 76 trains use the rubber-tyred metro technology developed by the RATP for some lines of the Paris Métro. In 2012, the Marseille Metro carried approximately 76.7 million passengers, making it a core part of the transport network in the Marseille urban area, with 49% of journeys using the metro. Since 1986, the Régie des transports métropolitains (Régie des transports de Marseille until 2016) has operated the network, operating it since it changed it name on behalf of the Aix-Marseille-Provence Metropolis. Timeline {| class="wikitable" ! width="15%" | Date ! width="10%" | Line ! width="75%" | Event |- | 26 November 1977 || align=center rowspan=2 | Line 1 || First section opened between La Rose and Saint-Charles |- | 11 March 1978 || Extended southwards from Saint-Charles to Castellane |- | 3 March 1984 || align=center rowspan=3 | Line 2 || Section opened between Joliette and Castellane |- | 1 February 1986 || Extended southwards from Castellane to Sainte-Marguerite Dromel |- | 14 February 1987 || Extended northwards from Joliette to Bougainville |- | 5 September 1992 || align=center rowspan=2 | Line 1 || Extended eastwards from Castellane to La Timone |- | 5 May 2010 || Extended eastwards from La Timone to La Fourragère |- | 16 December 2019 || align=center rowspan=2 | Line 2 || Extended northwards from Bougainville to Gèze |- | Before 2030 || Planned extension eastwards from Sainte-Marguerite Dromel to Saint-Loup Pagnol with 6 new stations, between and |- | After 2030 || align=center | Line 3 |Planned opening between Luminy and Saint-Charles with 15 new stations, between and |} History The first plans for a metro system in Marseille appeared in the early years of the 20th century, following the opening of the Paris metro. Many plans were put forward, but quickly abandoned due to lack of financing. The most serious proposal emanated in 1918 from the Compagnie d'électricité de Marseille, which proposed to build an underground network similar to the Paris métro. This proposal was met with fierce opposition from the Compagnie générale française de tramways, which owned and operated the city's tramway system. This project ultimately failed, and the idea of building a metro in Marseille was abandoned for many decades. The tramway system, badly damaged during the Second World War, was almost completely scrapped during the 1950s and replaced by buses (with the exception of line 68). However, by 1960, the city was suffering from severe congestion due to the growth in automobile use. New metro projects resurfaced in the mid-1960s, as a means to alleviate traffic congestion. After several years of studies, the city council voted unanimously in 1969 for the creation of a metro system. Construction of the first line started on August 13, 1973 and lasted until early 1977. Revenue operation started on November 26, 1977 on a portion of the line, between La Rose and Saint-Charles. The rest of the line opened on March 11, 1978. The plans for the second line were approved in 1978. Construction began in 1980. The central portion of the line, between Joliette and Castellane, opened on March 3, 1984. Southern and northern portions of the line were opened in February 1986 and February 1987 respectively. Subsequent extensions took place in the following years. </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Line 1 was first extended between Castellane and La Timone on September 5, 1992 (1.5 km, 2 new stations), and then between La Timone and La Fourragère (2.5 km, 4 new stations) in 2010. On 16 December 2019, line 2 was extended northwards to Bougainville to Gèze (0.9 km, 1 new station). Current network Map Rolling stock The rolling stock comprises 36 4-car trains, named MPM 76. Trains have a capacity of 472 passengers (including 182 seats). MPM 76 trains use the rubber tyre metro technology developed by the RATP for the Paris métro. Trains were built in Valenciennes, France, by a group of French companies which are now part of Alstom group. A first batch of 21 3-car trains was delivered in 1976, for line 1. A second batch of 15 was delivered in 1983, for line 2. In 1985, a fourth car was added on every train, in order to increase capacity. Commercial operation The metro system is operated by the Régie des Transports Métropolitains, on behalf of the Urban Community of Marseille Provence Métropole, which owns the infrastructure as well as the rolling stock. Service is open every day, from 5 am to 1 am the next day. Trains run every 3 minutes during rush hour, and every 10 minutes during evenings. The metro system transported 76.7 million passengers in 2012, leading to an average daily ridership of over 210,000. Planned developments A long extension of line 2 to Capitaine Gèze is expected to open in 2019, north of the current terminus station Bougainville. The new Capitaine Gèze'' station will feature a bus station and a park and ride facility. This short extension will reuse existing service tracks that currently lead to the Zoccola depot. The cost is estimated to be 80 million euros. Several other long-term extensions, including a southern extension of line 2 from Sainte-Marguerite to St-Loup, are being considered. In 2012, it was projected that the MPM76 rolling stock would be replaced starting in 2020, but there have been no further plans to do this since. See also List of Marseille Metro stations Marseille tramway List of metro systems References Bibliography External links Marseille Metro Map on Google earth with geolocation RTM – official website Marseille at UrbanRail.net Category:Rapid transit in France Category:Transport in Marseille Category:Rubber-tyred metros Category:Marseille Metro </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Slotermeer Slotermeer is a neighborhood of Amsterdam, Netherlands. It is divided in Slotermeer Northeast and Slotermeer Southwest. The center of Slotermeer is located in the area surrounding Plein '40-'45. Category:Neighbourhoods of Amsterdam </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> C-class destroyer (1943) The C class was a class of 32 destroyers of the Royal Navy that were launched from 1943 to 1945. The class was built in four flotillas of 8 vessels, the "Ca", "Ch", "Co" and "Cr" groups or sub-classes, ordered as the 11th, 12th, 13th and 14th Emergency Flotillas respectively. The sub-class names are derived from the initial 2 letters of the member ships' names, although the "Ca" class were originally ordered with a heterogeneous mix of traditional destroyer names. A fifth flotilla, the "Ce" or 15th Emergency Flotilla, was planned but were cancelled in favour of the s after only the first two ships had been ordered. The pennant numbers were all altered from "R" superior to "D" superior at the close of World War II; this involved some renumbering to avoid duplications. Design They were built as part of the War Emergency Programme, based on the hull and machinery of the pre-war J class, incorporating whatever advances in armament and naval radar were available at the time. Some of the class were completed in time for wartime service. All ships used the Fuze Keeping Clock High Angle Fire Control Computer. The "Ca" flotilla were generally repeats of the preceding Z class, and as such had a main gun armament of four QF Mk IV guns on Mk V mounts, which could elevate to 55 degrees to give an anti-aircraft capability. Close-in anti-aircraft armament generally consisted of two 40mm Bofors guns in a twin stabilized Hazemayer mount, supplemented by four single 2-pounder "pom-pom" anti aircraft guns on power operated mounts. Caprice differed in having a quadruple 2-pounder "pom-pom" instead of the Hazemayer Bofors mount, while Cassandra had eight Oerlikon 20 mm cannon instead of the single pom-poms. Torpedo armament consisted of eight torpedoes in two quadruple mounts, while 70 depth charges could be carried. The succeeding "Ch", "Co" and "Cr" flotillas were fitted with the new Mk VI HA/LA Director instead of the Mk I Type K director of the Z and Ca classes, while remote power control (RPC) gunlaying equipment was fitted. The additional weight of the new fire control equipment and the powered mountings for the 4.5 inch guns meant that only one quadruple torpedo mount was fitted, and the depth charge armament was reduced to 35 depth charges. Most of the ships were fitted with a single Hazemayer Bofors mount, although some of the later ships instead had the lighter and simpler Mk V twin Bofors mount. This was normally supplemented by two power operated single pom-pom mounts and two 20 mm Oerlikon cannon. They also introduced the all-welded hull into Royal Navy destroyer construction, beginning in Contest, with the "Cr" flotilla all being of all-welded construction. Late delivery of the Mk VI directors delayed completion such that all but one of the "Ch"s, "Co"s or "Cr"s entered service after the end of the Second World War. Only Comet was commissioned before VJ Day, in June 1945, albeit too late to see action. Caprice was the last destroyer built for the Royal Navy to be fitted with the ubiquitous quadruple QF 2 pounder "pom-pom" mounting Mark VII. The "Ca" flotilla were reconstructed in the late 1950s and early 1960s to be modernised for anti-submarine warfare and to serve as fast fleet escorts. One bank of torpedo tubes and one 4.5 in gun was removed, allowing two Squid triple-barreled anti-submarine mortars to be fitted, while the ships' obsolete gun Mk I Type K director was replaced by a modern Mk 6M director as fitted to Royal Navy frigates, and the remaining 4.5 in guns </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> fitted with RPC. Close in anti-aircraft armament was standardised as a single Mk V twin and two single 40 mm Bofors mounts. The ships were also fitted with new bridges; the post-refit bridge differed between the first four conversions (Cavendish, Carron, Cavalier and Carysfort), with open bridges and the later four (Caprice, Cassandra, Caesar and Cambrian) which were given frigate-type enclosed bridges. The remaining "Ch", "Co" and Cr" ships in the Royal Navy were given a less extensive modernisation, with one 4.5 in gun being replaced by twin Squids, modified fire control and a close in anti aircraft armament of 1 twin and four single Bofors guns. Chieftain, Chaplet and Comet were fitted as minelayers. Engineering The class were all fitted with two Admiralty 3-drum boilers with a pressure of at . All had Parsons single-reduction geared turbines, generating at 350 RPM, and driving the two shafts to produce a maximum of ( under full load condition). All were engined by their builders except Cossack and Constance, which were engined by Parsons. Their bunkers could hold 615 tons of oil fuel, giving them a radius of at and at . Ships * = flotilla leaders "Ca" (or 11th Emergency) Flotilla This flotilla was authorised under the 1941 Programme. The first pair was ordered from Yarrow on 16 February 1942; the other six were ordered on 24 March, a pair each from John Brown, Scotts and Cammell Laird. However, on 12 August 1942 the contract for the last pair was moved from Cammell Laird to White. Their originally-allocated names were altered to new names beginning with "Ca-" in November 1942. The John Brown pair - Caesar and Cavendish - were fitted as Leaders. On completion they formed the 6th Destroyer Flotilla for service in the Home Fleet. At the end of the war in Europe the flotilla was transferred to the East Indies Fleet and the ships arrived on station between August and November 1945, too late to see service against Japan. They remained in the Indian Ocean until May 1946 when they returned home and paid off into operational reserve. "Ch" (or 12th Emergency) Flotilla Six destroyers, the first of 26 'Intermediate' destroyers to be authorised under the 1942 Programme, were ordered on 24 July 1942, a pair each from Thornycroft, Scotts and Alexander Stephen. The fourth pair was originally intended to be ordered from Vickers Armstrongs, Walker-on-Tyne, but instead were ordered from Denny on 30 July. The Chequers and Childers were fitted as Leaders. "Co" (or 13th Emergency) Flotilla The first four of these destroyers were ordered in August 1942 - Comus and Concord on 7th, Contest on 12th and Consort on 14th. The remaining four destroyers were ordered on 12 September; Constance and Cossack were fitted as Leaders. "Cr" (or 14th Emergency) Flotilla All eight destroyers were ordered on 12 September 1942, two each from John Brown, Yarrow, White and Scotts; the John Brown pair - Crescent and Crusader - were fitted as Leaders. "Ce" (or 15th Emergency) Flotilla Two ships of this putative flotilla, the last of the 26 "Intermediate"-type destroyers authorised under the 1942 Programme, were ordered on 3 February 1942 from White. These two ships were to be named Centaur and Celt. However, with the decision to introduce a fresh design of Intermediate destroyer (which became the design), the White orders were amended to the new design and the names of the two ships were altered to Tomahawk and Sword respectively. Tomahawk was subsequently renamed again, becoming Scorpion, while Sword was finally cancelled on 15 October 1945. Image gallery See also List of ships of </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> the Second World War List of ship classes of the Second World War References Notes Footnotes Publications Maurice Cocker, Destroyers of the Royal Navy, 1893-1981, Ian Allan: London, 1981. Mike Critchley, British Warships Since 1945: Part 3: Destroyers, Maritime Books: Liskeard, UK, 1982. . Norman Friedman, British Destroyers & Frigates: The Second World War and After, Chatham Publishing, Robert Gardiner and Roger Chesneau, Conway's All The World's Fighting Ships 1922–1946, Conway Maritime Press: London, 1980. Robert Gardiner and Stephen Chumbley, Conway's All The World's Fighting Ships 1947–1995, Naval Institute Press: Annapolis, Maryland, USA, 1995. . David Hobbs, C Class Destroyers, Maritime Books: Liskeard, UK, 2012. Peter Hodges and Norman Friedman, Destroyer Weapons of World War 2, Naval Institute Press: Annapolis Maryland, USA, 1979. . H. T. Lenton, British and Empire Warships of the Second World War, Greenhill Books, H. T. Lenton, Navies of the Second World War: British Fleet & Escort Destroyers Volume Two, Macdonald: London, 1970. George Moore, Building for Victory: The Warship Building Programmes of the Royal Navy 1939 - 1945, World Ship Society, M.J. Whitley, Destroyers of World War Two: An International Encyclopedia, Cassell and Co.: London, 2000. . External links Category:Destroyer classes Category:Ship classes of the Royal Navy </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Wietersheim Wietersheim may refer to the following: Places Wietersheim, a district in the town of Petershagen, Germany. Persons Gustav Anton von Wietersheim (1884 – 1974), German General Walter von Wietersheim (1917), German Major Wend von Wietersheim (1900 – 1975), German Lieutenant General </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Ambrósio I of Kongo Ambrósio I Nimi a Nkanga was a mwenekongo of the Kingdom of Kongo who ruled from (March 1626 to March 7, 1631.) Rise to Power Ambrósio I was the nephew of Álvaro III and as such was a member of the royal House of Kwilu. When Alvaro III died on May 4 of 1622, he had only a young son to leave as heir. Instead of putting an easily manipulated juvenile on the throne at a time when the Portuguese under the renegade governor João Correia de Sousa was threatening the country, the royal council elected the Duke of Mbamba as King Pedro II. This ushered in the short-lived royal House of Nsundi. King Pedro II was peacefully succeeded by his son Garcia I but the calm was not to last. In 1626, Garcia was overthrown by disgruntled nobles led by Manuel Jordão the Duke of Nsundi. At the request of the royal ladies of the court, many of whom fervent partisans of the House of Kwilu, Jordão had Ambrósio crowned as king restoring the Kwilu kanda to power. Reign as King King Ambrósio fell out with his benefactor after two years and accused him of eyeing the throne. The king quickly moved to and succeeded at removing Jordão from his post as Duke of Nsundi, having him exiled to an island on the Congo River. This would not be the end of Ambrósio's troubles, as his reign was beset by rumors of conspiracy and war mobilizations which culminated in a massive revolt. King Ambrósio I was overthrown and killed March 7, 1631. References See also List of rulers of Kongo Kingdom of Kongo House of Kwilu Category:Manikongo of Kongo Category:1631 deaths Category:17th-century African people Category:Year of birth unknown </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Mehdi Jami Mehdi Jami (born February 1961 in Mashhad), is an Iranian journalist. Jami holds a master's degree in Persian language and literature from Ferdowsi University (1992). He is also a self taught photographer and filmmaker. His research interests include Iranian Studies (his book on Iranian Ancient Literary History published in Tehran, 2000) and Islamic History (he has written for Islamica Encyclopedia, Leiden 2008) and also Persian culture in Central Asia (Tajikistan and Uzbekistan). His documentary film on Tajik music Falak was released in 2005. Jami is a member of Artists Without Frontiers. From June 2006 to October 2008, he was the director of the Amsterdam-based Persian-language Radio Zamaneh which has been the first Persian media outlet based on the idea of Citizen Journalism. He worked for BBC's Persian Service between May 1996 and May 2006. His Persian weblog "Sibestan" (first post: May 2003) is one of the most read Persian weblogs. References اولین پستهای سیبستان First posts of Sibestan (in Persian) فهرست مقالات در دایره المعارف بزرگ اسلامی Entries of IE written by him (in Persian) پاره ای از فیلم چرخ و فلک Charkh-O-Falak, his documentary on Tajik music, a Fragment (in Tajiki and English) Auto-chronology of his life and works (in Persian) See also Persian culture Intellectual movements in Iran Category:1961 births Category:Living people Category:Iranian journalists Category:Iranian bloggers Category:People from Mashhad </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Patriotic Movement for National Rebirth Patriotyczny Ruch Odrodzenia Narodowego (PRON, ) was a Polish popular front that ruled the Polish People's Republic. It was created in the aftermath of the martial law in Poland (1982). Gathering various pro-communist and pro-government organizations, it was supposed to show unity and support for the government and Polish United Workers' Party (PZPR). PRON was created in July 1982 and dissolved in November 1989. Members The Patriotic Movement for National Rebirth included the following member parties: It also included the following organizations when founded: Stowarzyszenie PAX (Pax) Chrześcijańskie Stowarzyszenie Społeczne (ChSS) Polski Związek Katolicko-Społeczny (PZKS) It was later joined by many other organizations, such as All-Poland Alliance of Trade Unions, Związek Harcerstwa Polskiego and Towarzystwo Przyjaciół Dzieci. Few members of those organizations were aware of their membership in PRON, as PRON members included organizations, not individuals. PRON membership was required by communist propaganda, and necessary for any organization that wanted to exist on political scene with support of the government. PRON was mentioned in the amended Polish communist constitution, where it replaced the Front of National Unity. Like its predecessor, it was dominated by the PZPR; the minor parties had to accept the PZPR's "leading role" as a condition of their continued existence. PRON's chairman was the writer Jan Dobraczyński. The foundation committee included in addition to Dobraczyński: Marian Orzechowski, Janusz Reykowski, Andrzej Przypkowski, Edmund Męclewski, Jan Majewski, Andrzej Elbanowski, Józef Chlebowczyk, Władysław Ogrodziński, Walenty Milenuszkin, Wiesław Nowicki, Jerzy Stencel, Elżbieta Ciborowska, Jerzy Kejna, Piotr Perkowski, Józef Kiełb, Jerzy Ozdowski, Stanisław Rostworowski, Gizela Pawłowska, Zbigniew Gertych, Klemens Krzyżagórski, Anatola Klajna and Zbigniew Siatkowski. It was the sole organisation to put forward candidates in the 1985 election, which proved to be the last elections in which no opposition candidates were permitted to run. As such, it won every seat in the Sejm. References Patriotyczny Ruch Odrodzenia Narodowego entry at PWN Encyklopedia and - entries at WIEM Encyklopedia Category:Popular fronts of communist states Category:1982 establishments in Poland Category:Organizations established in 1982 Category:Organizations disestablished in 1989 Category:1980s in Poland Category:Polish United Workers' Party Category:Defunct political party alliances in Poland </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> T Cygni T Cygni is a binary star system in the northern constellation of Cygnus. It is a faint system but visible to the naked eye with a combined apparent visual magnitude of 4.93. Based upon an annual Parallax shift of , it is located 387 light years away. It is moving closer to the Earth with a heliocentric radial velocity of −24 km/s. The primary, component A, is a variable star, most likely of the slow irregular type, which ranges in magnitude from 4.91 down to 4.96. It is a giant star with a stellar classification of K3 III, which indicates it has exhausted the hydrogen at its core and evolved away from the main sequence. The measured angular diameter of this star, after correction for limb darkening, is . At the estimated distance of this star, this yields a physical size of about 27 times the Sun's radius. It is radiating 333 times the Sun's luminosity from its enlarged photosphere at an effective temperature of 4,190 K. The secondary companion, component B, is a magnitude 10.03 star located at an angular separation of along a position angle of 120°, as of 2012. In 1877 it was separated by with nearly the same position angle (121°). References Category:K-type giants Category:Slow irregular variables Category:Cygnus (constellation) Category:Durchmusterung objects 198134 102571 7956 Cygni, T </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Elmutasem Abushnaf Elmutasem Abushnaf (born 14 November 1991) is a Libyan footballer. International career International goals Scores and results list Libya's goal tally first. Honours Libya Winner African Nations Championship: 2014 References External links Category:1991 births Category:Living people Category:Libyan footballers Category:Libya international footballers Category:Association football forwards </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> David Morland IV David Morland IV (born April 3, 1969) is a Canadian professional golfer who has played on the Canadian Tour, PGA Tour, and Nationwide Tour. Morland was born in North Bay, Ontario. He turned professional in 1991 after going to Kent State University. He has played in 120 PGA Tour events and has three top-10 finishes. Professional wins (3) Nationwide Tour wins (2) 2002 Hibernia Southern Open 2003 SAS Carolina Classic Canadian Tour wins (1) 1999 McDonald's PEI Challenge See also 1999 PGA Tour Qualifying School graduates 2000 PGA Tour Qualifying School graduates 2003 Nationwide Tour graduates External links Category:Canadian male golfers Category:Kent State Golden Flashes men's golfers Category:PGA Tour golfers Category:Korn Ferry Tour graduates Category:Golfing people from Ontario Category:Sportspeople from North Bay, Ontario Category:1969 births Category:Living people </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Wyatt's Watchdogs Wyatt's Watchdogs is a 30-minute BBC1 situation comedy that starred Brian Wilde and Trevor Bannister. Created and written by Miles Tredinnick, the six-episode series was transmitted in the autumn of 1988. Alan J. W. Bell directed and the music was composed by Ronnie Hazlehurst. Although not initially written with him in mind, the series was developed as a vehicle for Wilde after he had left the cast of Last of the Summer Wine three years prior. Although initially getting passable ratings, the BBC felt that the sitcom had not really caught on, and it was dropped after one series; Wilde returned to Last of the Summer Wine in 1990. Overview The show was set in the fictional commuter village of Bradly Bush although actually filmed on location in Claygate, Surrey, England. Retired soldier Major John Wyatt (Brian Wilde) is spurred into action after his sister Edwina’s (Anne Ridler) home is burgled in broad daylight. Ignoring correct police procedures, he forms his own Neighbourhood Watch group of incompetents and patrols the streets in his Range Rover known locally as the ‘Dogmobile’. The Watch members are a hopeless bunch drawn together to fight crime, and bungling and personality clashes are highly evident. Brian Wilde as Major Wyatt clashes worst of all with Peter Pitt (Trevor Bannister), a smooth-talking womaniser and burglar-alarm salesman. He only has two things in mind - to flog his alarms and meet women! Other regular characters include the glamorous man-eater Virginia (April Walker), an interfering Vicar (David Jackson), and a frustrated police sergeant, Springer (James Warrior). The show actually had the working title of Every Street Should Have One but this was changed at the last minute to Wyatt's Watchdogs. The reason being that the BBC were transmitting the programme on Monday evenings straight after ITV's Coronation Street but before another BBC show called Streets Apart. Much of the series was devised to act as a vehicle for Wilde, who had left his role as Foggy Dewhurst in Last of the Summer Wine in 1985 over creative differences with Alan J. W. Bell. The character of Wyatt shared a number of similarities with Foggy, particularly with them both having a military background, although Wyatt was far more assertive and not afraid of conflict as Foggy was. Although it performed moderately in its time-slot, the BBC felt - who had hoped that, starring Wilde, it would have gained much of Last of the Summer Wines viewing figures - that the programme had not really caught on with viewers, and it was dropped after its initial series. However, with the production of this series, Wilde and Bell managed to settle their differences regarding Last of the Summer Wine, and the following year in 1989, when the next series of Summer Wine was due to start filming and actor Michael Aldridge having to leave the show for personal reasons, the character of Seymour Utterthwaite had to be replaced and Brian agreed to return as Foggy Dewhurst, staying with the series for seven more years. Episodes and cast Episode One: One Big, One Not So Big. Brian Wilde, Trevor Bannister, Anne Ridler, April Walker, James Warrior, David Jackson, Tom Radcliffe, Peter O'Sullevan (voice only), Julia Binstead and Sarah Whitlock. Episode Two: Getting Out And Spreading The Word. Brian Wilde, Trevor Bannister, Anne Ridler, April Walker, James Warrior, David Jackson, Noel Johnson, Lisa Bloor, Deborah Lavin, Frank Tregear, Richard Kane, Mary Blatchford, Ian Redford, Margaret Ashley, Helena McCarthy, Gabrielle Blunt and Aimée Delamain. Episode Three: Mark It Or Bust! (Directed by Andy Smith). Brian Wilde, Trevor Bannister, Anne Ridler, April Walker, </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> James Warrior, David Jackson, Eva Stuart, Timothy Carlton and Andrew Reardon. Episode Four: There Are Fairy Cakes At The Bottom Of My Garden. Brian Wilde, Trevor Bannister, Anne Ridler, April Walker, James Warrior, David Jackson, Robin Parkinson, Julie Morgan and Diana Fulker. Episode Five: Just Act Natural. Brian Wilde, Trevor Bannister, Anne Ridler, April Walker, James Warrior, David Jackson, Frederick Treves, Keith Smith, Russell Wootton, Richard Davies, Belinda Lee, Pamela Dale, Charles Appleby and Frank Tregear. Episode Six: A Clot On The Landscape. Brian Wilde, Trevor Bannister, Anne Ridler, April Walker, James Warrior, David Jackson, Clive Mantle, Claire Lacey, Martin Benson and Roger Ostime. DVD release To date, the series has not yet been released on DVD. External links Wyatt's Watchdogs site Official BBC Wyatt's Watchdogs archive site Wyatt's Watchdogs 2017 interview with Miles Tredinnick Category:BBC television sitcoms Category:1980s British comedy television series Category:1988 British television series debuts Category:1988 British television series endings Category:English-language television programs </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Easter orchid Easter orchid is the common name of one of three fragrant orchids: Cattleya mossiae, a native of Venezuela Cattleya schroederae, a native of Colombia Earina autumnalis, a native of New Zealand Plants named Easter orchid </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Telstra Corporation Ltd v Commonwealth Telstra Corporation Limited v The Commonwealth was an important case decided in the High Court of Australia on 6 March 2008. It concerned a dispute between Telstra Corporation Ltd and the Commonwealth of Australia, the Australian Competition and Consumer Commission (ACCC) and eleven other ISPs regarding the ownership of and shared access obligations to the Telstra controlled unconditional local loop (ULL) copper telecommunications network. The case determined that although Telstra owned and maintained this network, its ownership over these assets were always subject to a statutory access regime permitting other carriers rightful access these assets. It is an important case in Australia constitutional law in that it explored the interpretation of the nature of property and acquisition. More importantly, the ruling had a resounding effect on competition and consumer protection within the Australian telecommunications industry by upholding the right to fair competition of competing carriers, thus preventing a monopoly control of Internet and telephony services by Telstra. Background to the case From the 1880s Australia had developed an extensive public switched telephone network (PSTN), copper infrastructure which is widely dispersed throughout the country and immediately evident in most business and residential premises as the standard telephone network. This infrastructure is distributed nationwide by telephone exchanges, from which it locally extends into customer premises. The physical copper connection between the customer premises and the telephone exchange is known as a "local loop". There are about 10.1 million local loops and about 5,120 local exchanges. Local loops can carry services in two ways, unconditioned and shared. Unconditioned local loop services (ULLS) are "unconditioned" in that the electrical properties of the loop are not altered by any physically connected equipment. Local sharing services (LSS) are unconditioned local loops where one carrier uses the lower frequencies range of the line, known as the "voiceband" used in regular telephony, whilst that same carrier or another uses the higher frequency range of that same line for high speed services such as ADSL internet. Prior to Federation, the PSTN infrastructure that then existed was owned and maintained by the colonial governments. The Constitution provided for the transfer of the network to the Commonwealth of Australia, who operated the network through the Postmaster-General's Department. In July 1975, the assets constituting the PSTN were transferred to a body then called the Australian Telecommunications Commission, later changed to the Australian Telecommunications Corporation, created under the Telecommunications Act 1975 and began trading as a monopoly under the name Telecom Australia. In a Parliamentary move to invite competition, Telstra Corporation Limited, then known as the Australian and Overseas Telecommunications Corporation, was incorporated under the Corporations Law of ACT in November 1991, as a company limited by shares, the sole shareholder being the Australian Commonwealth. The PSTN was then vested in Telstra, as of 1 February 1992, in accordance with the Telecommunications Act 1991 of which one of the stated objects is "creating a regulatory environment for the supply of telecommunications services which promotes competition and fair and efficient market conduct". Once privatised, Telstra operation became subject to the Trade Practices Act 1974 which requires that competitors be given access to the PSTN and related services, and empowered The Australian Competition and Consumer Commission (ACCC) to make determinations about the terms and conditions of access obligations and the pricing principles applicable to services relating to the PSTN, and in this case ULLS and LSS. It was this enforced pricing model that Telstra rejected, claiming they are forced to provide wholesale access to competitors at below cost price, which they believed amounted to an acquisition of property other than </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> on just terms contrary to Section 51(xxxi) of the Constitution. Thus, Telstra began proceedings in the High Court of Australia against the Commonwealth, the ACCC, and eleven of its competitors, seeking adequate compensation and the ability to charge its competitors more for access. The eleven competitors involved were internet service providers Primus Telecommunications, Optus Networks, Chime Communications, XYZED, Powertel Limited, Request Broadband, NEC Australia, Macquarie Telecom, AMCOM, Adam Internet, and Agile. The Telecommunications Access Regime The PSTN was vested in Telstra, with effect from 1 February 1992, under the Telecommunications Act 1991 which commenced on 1 July 1991. This meant that all of its provisions had commenced prior to the PSTN actually being vested in Telstra. This legislation limited a former monopoly control over the telecommunications industry by promoting the fair competition subject to the Australian Telecommunications Access Regime set out in Part XIC of the Trade Practices Act. This regime applies to telecommunications services which, following a public inquiry, are declared by the ACCC as "declared services". Section 152AL(3) of the TPA provides for the declaration of these services subject to Part XIC. Once a service becomes a declared service the service provider that provides that service must then adhere to the standard access obligations as described in Section 152AR. Part XIC of the Trade Practices Act was inserted in that Act by the Trade Practices Amendment (Telecommunications) Act 1997. Argument Telstra rejected the regulation of ULLS and LSS under Part XIC of the Trade Practices Act, alleging that the wholesale access price caps imposed on it by the ACCC do not allow it to recover the operating costs of the actual service being accessed and retailed by a competitor, alleging that this amounted to a compulsory acquisition of Telstra property by the Commonwealth, and access-seeking competitors, on unjust terms contrary to section 51(xxxi) of the Constitution. Section 51(xxxi) of the Constitution states that the Parliament has legislative power to create laws relating to:the acquisition of property on just terms from any State or person for any purpose in respect of which the Parliament has power to make laws. Impugned provisions The argument made on Telstra's behalf focused directly on Sections 152AL(3) and 152AR of the Trade Practices Act, and asked whether these two Sections, in their applications to ULL and LSS, were beyond the legislative powers of parliament by way of Section 51(xxxi). Therefore, rejecting the validity of Sections 152AL(3) and 152AR which it claimed affect an unconstitutional acquisition of Telstra property by the Commonwealth on unjust terms. As mentioned earlier these two Sections provide for the declaration of services by the ACCC and impose standard access obligations, terms and conditions which include the monetary price bracket limiting what Telstra is allowed to charge its competitors for access to declared services. If then, as Telstra continued to argue, terms and conditions may be fixed for its compliance with standard access obligations in respect to local loops, ULL and LSS, which are terms that differ from those that would be fixed in arm's length (all parties being independent and equal) bargaining between it and the access seeker, the provisions that lead to that result provide for an acquisition of property otherwise than on just terms. Section 152EB Telstra submitted that this conclusion is not denied by the provisions of Section 152EB which deal expressly with the subject of just terms. If a Commonwealth law provides for the acquisition of property on unjust terms, then Section 51(xxxi) renders that law invalid, since it is beyond the legislative powers of the Commonwealth to create such a law, then Section 152EB, the </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> final provision of Part XIC of the Trade Practices Act, acts to save that law from invalidity, via reasonable compensation by the Commonwealth. Telstra argued that this provision did not save the impugned provisions from invalidity stating that relevant acquisitions "occur by the imposition of the standard access obligations on carriers or providers pursuant to Section 152AR" and not by any ACCC determination. Furthermore, any acquisition only comes into operation when a relevant request for the service is made by a provider, and not by any ACCC determination. Telstra stated that Section 152EB should be considered engaged only in respect to determinations that effect an acquisition, and that the only relevant determinations are those made by the ACCC. Thus, Telstra argued, this provision did not apply to any acquisition effected by Section 152AR since the relevant acquisition is effected by the provider's request for access engaging the standard obligations under Section 152AR, and not by any determination made by the ACCC. The High Court rejected Telstra's interpretation of Section 152EB by stating that the said provision refers to determinations that would "result in" an acquisition of property on unjust terms, and not to determinations that would "effect" such an acquisition. The High Court also pointed to the fact that Telstra's argument involves a dispute about the terms and conditions of complying with the service access obligations fixed by ACCC determinations, thus enabling the operation of Section 152EB. As a result of the High Court's position, Telstra next suggested that the impugned provisions do effect an acquisition of property, and just terms must then be afforded by the operation of Section 152EB, removing any claim of invalidity so long as the Commonwealth pays reasonable compensation. At this stage the High Court found it important to consider the nature of the "acquisition of property". Acquisition of property? In Australian courts Section 51(xxxi) of the Constitution focuses on matters of substance rather than form, and therefore the definitions of "acquisition" and "property" are to be interpreted liberally. In the case of Yanner v Eaton. it was mentioned that the word "property" is often used to refer to something that belongs to another, though in law, "property" does not refer to a thing; it is a description of a legal relationship with a thing. It refers to a degree of power that is recognised in law as power permissibly exercised over the thing. The concept of "property" may be elusive. Usually it is treated as a "bundle of rights". In this same case it was also accepted that "an extensive frame of reference is created by the notion that “property” consists primarily in control over access. Much of our false thinking about property stems from the residual perception that “property” is itself a thing or resource rather than a legally endorsed concentration of power over things and resources." Such considerations were vital to the matter concerning Telstra, as they submitted that the physical disconnection of the local loop from its equipment and its physical connection to another carrier’s equipment constituted an acquisition of property. Due to these physical acts, Telstra claimed it lost "control of and the ability to use the infrastructure" which it "owned". The competing carriers, listed as defending parties, chose to shift the argument from the physical steps in connecting and disconnecting ULL and LLS to that of whether there had ever been any actual "possession" on their part, since Telstra installs, repairs, and maintain the network at all times. All connections and disconnections of ULL and LLS are carried out by Telstra, regardless of whether those services relate to its </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> own customers or customers of competing carriers. These same defendants pointed to the fact that end-users, the customers, can and do chose which service provider connects equipment to the local loops entering their premises, and as long as that choice is not Telstra, then Telstra have no use for that local loop. The defendant's arguments sought to invoke certain elements of the long line of cases in the Australian High Court in which Section 51(xxxi) has been considered. In addressing these arguments the Court decided that rather than begin from the classification of rule and its exceptions, it would instead recognise observations made by Brennan CJ, Toohey, Gaudron, McHugh and Gummow JJ in Victoria v The Commonwealth (Industrial Relations Act Case): "It is well established that the guarantee effected by s 51(xxxi) of the Constitution extends to protect against the acquisition, other than on just terms, of “every species of valuable right and interest including ... choses in action”. It has been held to prohibit the extinguishment of vested causes of action. At least that is so if the extinguishment results “in a direct benefit or financial gain ... and the cause of action is one that arises under the general law”." Judgment The High Court held that there were three key factors on the history of the Australian communications industry that led to the conclusion that operation of the access regime did not result in an acquisition of property on unjust terms. Firstly the PSTN (including the local loops in dispute) which Telstra now "owns" was originally owned by the Commonwealth itself. Secondly, prior to the vesting of the PSTN (and local loops) in Telstra's predecessor, the Telecommunications Act 1991 had commenced on 1 July 1991, establishing the access regime that gave competitors the rights to interconnect to the PSTN and to obtain access to services it supplies. Thirdly, at the time the PSTN was vested in Telstra, Telstra was wholly owned by the Commonwealth. And so it was maintained that when the PSTN was transferred from the Commonwealth to Telstra's predecessor, Telstra's right over the network were always subject to the access rights of its competitors to its network and services. Thus, the High Court unanimously dismissed Telstra's case holding that Sections 152AL(3) and 152AR of the Trade Practices Act were not invalid and that Telstra never actually owned any of the assets that comprise the PSTN except under and in accordance with the legislative provisions for "promoting ... competition in the telecommunications industry generally and among carriers". Furthermore, it was stated that Telstra's argument was "synthetic and unreal" because it was based on an assumption that Telstra has greater rights over the PSTN than it actually has. Consequences In a media release on its website, dated 8 March 2008, Telstra announced that it accepted the High Court's ruling and stood by its decision to challenge the ACCC's powers under Section 51(xxxi) of the Constitution, though believes Australia lost an opportunity for further broadband investment when the regulatory regime that rewards competitors who "ride on Telstra’s network rather than invest in Australia" was upheld. Other commentators believed this outcome to be a clear rebuttal to Telstra's "obsession with privatisation for its own sake" and protection of its remaining monopoly under the direction of CEO at the time Sol Trujillo, a United States telecommunications executive with a history of aggressive litigation towards the US government and its regulators, who was hired by the Telstra board for his ability to uphold a "culture of litigating in order to protect its monopoly privileges". Had such privileges as those relating to ULL </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> and SSL been upheld, Telstra would be able to charge monopoly-inflated prices for access to its network, thus reducing competition and increasing consumer access fees. Exploitation of such monopoly control would result in increased access costs for competitors, their private and business customers, and eventually, all Australian online services. After Trujillo's resignation from Telstra in February 2009, new CEO David Thodey reached out to the Australian Government and regulators such as the ACCC in an effort to repair the relationships damaged during his predecessor's reign. The taking down of its propaganda website nowweretalking.com.au, which the company used to criticise regulators and competitors, marked the removal of one of the final vestiges of the Trujillo era. References Category:High Court of Australia cases Category:Acquisition of property in the Australian Constitution cases Category:2008 in Australian law Category:2008 in case law Category:Telstra </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Jewels of Desire Jewels of Desire is a 1927 silent film directed by Paul Powell and starring Priscilla Dean. It was released through Producers Distributing Corporation. A print is preserved at the UCLA Film and Television Archive. Cast Priscilla Dean - Margarita Solano John Bowers - Maclyn Mills Walter Long - Pedro Luke Cosgrove - Captain Blunt Syd Crossley - Taxi driver Ernie Adams - The Rat Raymond Wells - Spanish Joe Marie Percivale - Old Indian Woman References External links Jewels of Desire at IMDb.com allmovie/synopsis: Jewels of Desire Category:1927 films Category:American silent feature films Category:American films Category:Films directed by Paul Powell (director) Category:Films based on short fiction Category:1920s romance films Category:American black-and-white films Category:Producers Distributing Corporation films </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Inverse copular constructions In linguistics, inverse copular constructions, named after Moro (1997), are a type of inversion in English where canonical SCP word order (subject-copula-predicative expression, e.g. Fred is the plumber) is reversed in a sense, so that one appears to have the order PCS instead (predicative expression-copula-subject, e.g. The plumber is Fred). The verb in these constructions is always the copula be (am, are, is, was, were). Inverse copular constructions are intriguing because they render the distinction between subject and predicative expression difficult to maintain. The confusion has led to focused study of these constructions, and their impact on the theory of grammar may be great since they appear to challenge the initial binary division of the sentence (S) into a subject noun phrase (NP) and a predicate verb phrase (VP) (S → NP VP), this division being at the core of all phrase structure grammars (as opposed to dependency grammars, which do not acknowledge the binary division). Examples Inverse copular constructions involve nouns and noun phrases, but they do not allow the post-copula nominal to be a personal pronoun: a. The cause of the riot is a picture on the wall. b. A picture on the wall is the cause of the riot. c. *A picture on the wall is it. a. Fred is the plumber. b. The plumber is Fred. c. *The plumber is he. The defining trait of the inverse copular constructions is that two counts of inversion appear to have occurred: the normal subject has inverted to a post-verb position, and the predicative nominal has inverted to the pre-verb position. The verb is a finite form of the copula 'be' (am, are, is, was, were). This type of inversion is generally NOT possible with other verbs. Subject-verb agreement Inverse copular constructions where the inverted predicative expression is a noun phrase are noteworthy in part because subject-verb agreement can (at least in English) be established with the pre-verb predicative NP as opposed to with the post-verb subject NP, e.g. a. The pictures are a problem. b. A problem is/??are the pictures. a. Those kids are an annoyance. b. An annoyance is/??are those kids. In the inverse copular constructions, the copula agrees with the singular predicative expression to its left as opposed to with the plural subject to its right. This phenomenon seems to be limited to English (and possibly French); it does not occur in related languages such as German, e.g. a. Die Bilder sind ein Problem. 'The pictures are a problem.' b. Ein Problem sind/*ist die Bilder. 'A problem are/is the pictures.' Nor does it occur in some Romance languages, e.g. Italian: a. Queste foto sono la causa della rivolta. 'These photos are the cause of the revolt.' b. La causa della rivolta sono/*è queste foto. 'The cause of the revolt are/is the photos.' The fact that English (unlike German and Italian) demands subject-verb agreement to occur with the pre-verb NP generates confusion about what should qualify as the subject NP. From a morphological point of view, the pre-verb NP in inverse copular constructions should count as the subject, but from the perspective of information structure (e.g. definiteness, old information, specificity), the post-verb NP should be the subject. Importance for the theory of grammar Inverse copular constructions challenge one of the major dogmas of the theory of clause or sentence structure, i.e. that the two basic constituents of a sentence - the noun phrase (NP) and the verb phrase (VP) - are associated with the logical/grammatical functions of subject and predicate (cf. phrase structure rules and sentence). In fact, copular sentences that maintain the canonical </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> groupings are not adequate on empirical grounds, since a very unorthodox left-branching structure is necessary, or if one rejects the canonical groupings and positions the subject inside a VP-like constituent, then one has to assume that the subject NP and copula verb can form a type of VP to the exclusion of the predicative expression. See also Copula Dependency grammar Discontinuity Inversion Subject-auxiliary inversion Phrase structure grammar Predicate Predicative expression Subject-verb inversion Notes References Heycock, C. and A. Kroch 1998. Inversion and equation in copular sentences. In A. Alexiadou et al. (eds) ZAS Papers in Linguistics 10. 71 - 87. Zentrum für Allgemeine Sprachwissenschaft, Berlin. Mikkelsen, L. 2005. Copular clauses: Specification, predication, and equation. Linguistics Today 85. Amsterdam: John Benjamins. Moro, A. 1997. The raising of predicates. Predicative noun phrases and the theory of clause structure, Cambridge Studies in Linguistics, Cambridge University Press, Cambridge, England. Pereltsvaig, A. 2001. Copular sentences and the architecture of grammar, Ph.D. Thesis, McGill University, Canada. Category:Syntax Category:Syntactic entities Category:Syntactic transformation </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Joseph F. Loy Joseph F. Loy (1824 – January 29, 1875) was a lawyer and politician. Born in Somerset County, Pennsylvania, Loy went to school in Pennsylvania. He studied law and was admitted to the Pennsylvania bar. In 1850, Loy moved to De Pere, Wisconsin and then to Green Bay, Wisconsin where he practiced law. In 1853 and 1854, Loy served in the Wisconsin State Senate. During the American Civil War, Loy served in the 4th Wisconsin Volunteer Cavalry Regiment and was commissioned a captain. Notes Category:1824 births Category:1875 deaths Category:Politicians from Green Bay, Wisconsin Category:People from Somerset County, Pennsylvania Category:People of Wisconsin in the American Civil War Category:Pennsylvania lawyers Category:Wisconsin lawyers Category:Wisconsin state senators Category:19th-century American politicians </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Ranganna Ranganna is a 1997 Indian Kannada language romantic action drama written and directed by H. Vasu. The film stars Jaggesh as an auto driver, Ranga, who tries to mend the broken relationship of his uncle with the help of Roopa by reuniting him with his stubborn wife who happens to be Roopa's mother. Vijayalakshmi plays Ranga's love interest, Roopa. Srinath, Kavitha, Dheerendra Gopal and Loknath essayed other important roles. Rajesh Ramanath composed the soundtrack whileJ. G. Krishna handled the cinematography. The film marks the first collaboration of director H. Vasu with music director Rajesh Ramanath and second collaboration with Jaggesh and producer Sa Ra Govindaraju after Bhanda Alla Bahaddur which was released in the same year. The film received a U certificate from the CBFC without any cuts on 7 October 1997. Sri Ganesh Videos acquired the video rights and released and marketed the film in VCD and DVD formats. Cast Jaggesh Vijayalakshmi Srinath Kavitha Loknath Dheerendra Gopal Harish Rai Honnavalli Krishna Bank Janardhan M. S. Karanth Soundtrack The soundtrack album comprises 5 songs composed by Rajesh Ramanath. The audio rights of the film was sold to Lahari Music. Auto Auto — sung by S. P. Balasubrahmanyam ; lyrics by S. R. Shastry Bandlu Bandlu — sung by Rajesh Krishnan and Soumya ; lyrics by Shyamsundar Kulkarni Baaro Baaro — sung by Rajesh Krishnan and K. S. Chithra ; lyrics by Doddarange Gowda Mangoli Mangol — sung by Rajesh Krishnan and Soumya ; lyrics by K. Kalyan Aajare Aaja — sung by Rajesh Krishnan and K. S. Chithra ; lyrics by Sri Ranga References External links at Gaana Category:1997 films Category:1990s Kannada-language films Category:Indian films </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Prahok Prahok (ប្រហុក) is a crushed, salted and fermented fish paste (usually of mudfish) that is used in Cambodian cuisine as a seasoning or a condiment. It originated as a way of preserving fish during the months when fresh fish was not available in abundant supply. Because of its saltiness and strong flavor, it was used as an addition to many meals in Cambodian cuisine, such as soups and sauces. A Cambodian saying goes, "No prahok, no salt", referring to a dish that is of poor flavor or bland thus highlighting its essentiality in Cambodian cuisine. Prahok has a strong and distinct smell, earning the nickname "Cambodian cheese" among food writers. Prahok is usually eaten as a main course with white rice and vegetable such as yardlong bean, cucumbers, and Thai eggplant. Prahok is sometimes distributed as a donation to victims of flood or drought by charities and other organizations. It can be eaten cooked or fried, but is usually not eaten raw because of health issues (raw Prahok cannot be stored long due to spoilage if not consumed in a short period) and the unpleasant smell it makes. Varieties and production Prahok is made with various fish and methods of fermentation. Fish used include mudfish (Channa spp.) and moonlight gourami (Trichogaster microlepis). One noted variety made with a gourami species is called Prahok Kanthara and is attributed to a Laotian style of preparation. Prahok is obtained by crushing or grinding fresh fish after de-scaling, gutting and cleaning them. They can be crushed underfoot, like wine grapes, or processed by machine. After the fish is crushed, it is left in the sun for a full day, then salted. The prahok is fermented in large clay jars covered with a lid made of woven bamboo strips. Afterwards, the prahok can be eaten just after 20 days of fermentation, but the best quality Prahok is left to ferment longer, up to three years. Prahok is also produced in Vietnam and imported for Cambodian diaspora United States. Prahok dishes Prahok can be prepared and served in several different ways. Below are dishes where prahok is a main component. Fried prahok Prahok chien (ប្រហុកជៀន) It is usually mixed with meat (usually beef or pork) and chili peppers. It can also be eaten as a dip, accompanied by vegetables like cucumbers or eggplants, and rice. Covered prahok Prahok Kob or Prahok aing (ប្រហុកកប់) or (ប្រហុកអាំង) This type of prahok is covered with banana leaves and left to cook under pieces of rock beneath a fire or over the coals. Raw prahok Prahok chao (ប្រហុកឆៅ) This type of prahok can be used to make a paste with lemon grass, lime juice, fresh peppers, and eggplant eaten with (usually cooked rare) beef steak. Also this is the type of prahok preferably used as a dipping paste for vegetables and fruits. See also , Burmese fish paste , Filipino fish paste References External links LeisureCambodia.com - The Story of Prahok article. ThingsAsian.com - Got fish? It's Prahok season in Cambodia article. Category:Cambodian cuisine Category:Fish sauces Category:Umami enhancers Category:Food paste </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Heart development Heart development (also known as cardiogenesis) refers to the prenatal development of the human heart. This begins with the formation of two endocardial tubes which merge to form the tubular heart, also called the primitive heart tube, that loops and septates into the four chambers and paired arterial trunks that form the adult heart. The heart is the first functional organ in vertebrate embryos, and in the human, beats spontaneously by week 4 of development. The tubular heart quickly differentiates into the truncus arteriosus, bulbus cordis, primitive ventricle, primitive atrium, and the sinus venosus. The truncus arteriosus splits into the ascending aorta and pulmonary artery. The bulbus cordis forms part of the ventricles. The sinus venosus connects to the fetal circulation. The heart tube elongates on the right side, looping and becoming the first visual sign of left-right asymmetry of the body. Septa form within the atria and ventricles to separate the left and right sides of the heart. Early development The heart derives from embryonic mesodermal germ-layer cells that differentiate after gastrulation into mesothelium, endothelium, and myocardium. Mesothelial pericardium forms the outer lining of the heart. The inner lining of the heart, lymphatic and blood vessels, develop from endothelium. Endocardial tubes In the splanchnopleuric mesenchyme on either side of the neural plate, a horseshoe-shaped area develops as the cardiogenic region. This has formed from cardiac myoblasts and blood islands as forerunners of blood cells and vessels. By day 19, an endocardial tube begins to develop in each side of this region. These two tubes grow and by the third week have converged towards each other to merge, using programmed cell death to form a single tube, the tubular heart. From splanchnopleuric mesenchyme, the cardiogenic region develops cranially and laterally to the neural plate. In this area, two separate angiogenic cell clusters form on either side and coalesce to form the endocardial tubes. As embryonic folding continues, the two endocardial tubes are pushed into the thoracic cavity, where they begin to fuse together, and this is completed at about 22 days. At around 18 to 19 days after fertilisation, the heart begins to form. This early development is critical for subsequent embryonic and prenatal development. The heart is the first functional organ to develop and starts to beat and pump blood at around day 21 or 22. The heart begins to develop near the head of the embryo in the cardiogenic area. Following cell signalling, two strands or cords begin to form in the cardiogenic region As these form, a lumen develops within them, at which point, they are referred to as endocardial tubes. At the same time that the tubes are forming other major heart components are also being formed. The two tubes migrate together and fuse to form a single primitive heart tube, the tubular heart which quickly forms five distinct regions. From head to tail, these are the truncus arteriosus, bulbus cordis, primitive ventricle, primitive atrium, and the sinus venosus. Initially, all venous blood flows into the sinus venosus, and contractions propel the blood from tail to head, or from the sinus venosus to the truncus arteriosus. The truncus arteriosus will divide to form the aorta and pulmonary artery; the bulbus cordis will develop into the right ventricle; the primitive ventricle will form the left ventricle; the primitive atrium will become the front parts of the left and right atria and their appendages, and the sinus venosus will develop into the posterior part of the right atrium, the sinoatrial node and the coronary sinus. Heart tube position The central part of cardiogenic area is in </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> front of the oropharyngeal membrane and the neural plate. The growth of the brain and the cephalic folds push the oropharyngeal membrane forward, while the heart and the pericardial cavity move first to the cervical region and then into the chest. The curved portion of the horseshoe-shaped area expands to form the future ventricular infundibulum and the ventricular regions, as the heart tube continues to expand. The tube starts receiving venous drainage in its caudal pole and will pump blood out of the first aortic arch and into the dorsal aorta through its polar head. Initially the tube remains attached to the dorsal part of the pericardial cavity by a mesodermal tissue fold called the dorsal mesoderm. This mesoderm disappears to form the two pericardial sinuses the transverse and the oblique pericardial sinuses, which connect both sides of the pericardial cavity. The myocardium thickens and secretes a thick layer of rich extracellular matrix containing hyaluronic acid which separates the endothelium. Then mesothelial cells form the pericardium and migrate to form most of the epicardium. Then the heart tube is formed by the endocardium, which is the inner endothelial lining of the heart, and the myocardial muscle wall which is the epicardium that covers the outside of the tube. Heart folding The heart tube continues stretching and by day 23, in a process called morphogenesis, cardiac looping begins. The cephalic portion curves in a frontal clockwise direction. The atrial portion starts moving in a cephalic ally and then moves to the left from its original position. This curved shape approaches the heart and finishes its growth on day 28. The conduit forms the atrial and ventricular junctions which connect the common atrium and the common ventricle in the early embryo. The arterial bulb forms the trabecular portion of the right ventricle. A cone will form the infundibula blood of both ventricles. The arterial trunk and the roots will form the proximal portion of the aorta and the pulmonary artery. The junction between the ventricle and the arterial bulb will be called the primary intra-ventricular hole. The tube is divided into cardiac regions along its craniocaudal axis: the primitive ventricle, called primitive left ventricle, and the trabecular proximal arterial bulb, called the primitive right ventricle. This time no septum is present in heart. Heart chambers Sinus venosus In the middle of the fourth week, the sinus venosus receives venous blood from the poles of right and left sinus. Each pole receives blood from three major veins: the vitelline vein, the umbilical vein and the common cardinal vein. The sinus opening moves clockwise. This movement is caused mainly by the left to right shunt of blood, which occurs in the venous system during the fourth and fifth week of development. When the left common cardinal vein disappears in the tenth week only the oblique vein of the left atrium and the coronary sinus remain. The right pole joins the right atrium to form the wall portion of the right atrium. The right and left venous valves fuse and form a peak known as the septum spurium. At the beginning, these valves are large, but over time the left venous valve and the septum spurium fuse with the developing atrial septum. The upper right venous valve disappears, while the bottom venous valve evolves into the inferior valve of the vena cava and the coronary sinus valve. Heart wall The main walls of the heart are formed between day 27 and 37 of the development of the early embryo. The growth consists of two tissue masses actively growing that approach one another until they merge </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> and split light into two separate conduits. Tissue masses called endocardial cushions develop into atrioventricular and conotroncal regions. In these places, the cushions will help in the formation of auricular septum, ventricular conduits, atrio-ventricular valves and aortic and pulmonary channels. Atria At the end of the fourth week, a crest grows that leaves the cephalic part. This crest is the first part of the septum primum. The two ends of the septum extend into the interior of the endocardial cushions in the atrioventricular canal. The opening between the bottom edge of the septum primum and endocardial cushions is the ostium primum (first opening). The extensions of the upper and lower endocardial pads grow along the margin of the septum primum and close the ostium primum. Coalescence of these perforations will form the ostium secundum (second opening), which allows blood to flow freely from the right atrium to the left. When the right of the atrium expands due to the incorporation of the pole of the sinus, a new fold appears, called the septum secundum. At its right side it is fused with the left venous valve and the septum spurium. A free opening will then appear, called the foramen ovale. The remains of the upper septum primum, will become the valves of the foramen ovale. The passage between the two atrial chambers consists of a long oblique slit through which blood flows from the right atrium to the left. Ventricles Initially, a single pulmonary vein develops in the form of a bulge in the back wall of the left atrium. This vein will connect with the veins of the developing lung buds. As development proceeds the pulmonary vein and its branches are incorporated into the left atrium and they both form the smooth wall of the atrium. The embryonic left atrium remains as the trabecular left atrial appendage, and the embryonic right atrium remains as the right atrial appendage. Septum formation of the atrioventricular canal At the end of the fourth week, two atrioventricular endocardial cushions appear. Initially the atrioventricular canal gives access to the primitive left ventricle, and is separated from arterial bulb by the edge of the ventricular bulb. In the fifth week, the posterior end terminates in the center part of the upper endocardial cushion. Because of this, blood can access both the left primitive ventricle and the right primitive ventricle. As the anterior and posterior pads project inwardly, they merge to form a right and left atrioventricular orifice. Atrioventricular valves When forming intra-atrial septa, atrio-ventricular valves will begin to grow. A muscular interventricular septum begins to grow from the common ventricle to the atrio-ventricular endocardial cushions. The division begins in the common ventricle where a furrow in the outer surface of the heart will appear the interventricular foramen eventually disappears. This closure is achieved by further growth of the muscular interventricular septum, a contribution of trunk crest-conal tissue and a membranous component. Valves and outflow tracts Truncus septum formation and arterial cone The arterial cone is closed by the infundibular cushions. The trunk cones are closed by the forming of an infundibulotroncal septum, which is made from a straight proximal portion and distal spiral portion. Then, the narrowest portion of the aorta is in the left and dorsal portion. The distal portion of the aorta is pushed forward to the right. The proximal pulmonary artery is right and ventral, and the distal portion of the pulmonary artery is in the left dorsal portion. Pacemaker and conduction system The rhythmic electrical depolarization waves that trigger myocardial contraction is myogenic, which means that they begin in the heart </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> muscle spontaneously and are then responsible for transmitting signals from cell to cell. Myocytes that were obtained in the primitive heart tube, start beating as they connect together by their walls in a syncytium. Myocytes initiate rhythmic electrical activity, before the fusion of the endocardial tubes. The heartbeat begins in the region of the pacemaker which has a spontaneous depolarization time faster than the rest of myocardium. The primitive ventricle acts as initial pacemaker. But this pacemaker activity is actually made by a group of cells that derive from the sinoatrial right venous sinus. These cells form an ovoid sinoatrial node (SAN), on the left venous valve. After the development of the SAN, the superior endocardial cushions begin to form a pacemaker as known as the atrioventricular node. With the development of the SAN, a band of specialized conducting cells start to form creating the bundle of His that sends a branch to the right ventricle and one to the left ventricle. Most conduction pathways originate from the cardiogenic mesoderm but the sinus node may be derived from the neural crest. The human embryonic heart begins beating approximately 21 days after fertilization, or five weeks after the last normal menstrual period (LMP), which is the date normally used to date pregnancy in the medical community. The electrical depolarizations that trigger cardiac myocytes to contract arise spontaneously within the myocyte itself. The heartbeat is initiated in the pacemaker regions and spreads to the rest of the heart through a conduction pathway. Pacemaker cells develop in the primitive atrium and the sinus venosus to form the sinoatrial node and the atrioventricular node respectively. Conductive cells develop the bundle of His and carry the depolarization into the lower heart. Cardiac activity is visible beginning at approximately 5 weeks of pregnancy. The human heart begins beating at a rate near the mother’s, about 75-80 beats per minute (BPM). The embryonic heart rate (EHR) then accelerates linearly for the first month of beating, peaking at 165-185 BPM during the early 7th week, (early 9th week after the LMP). This acceleration is approximately 3.3 BPM per day, or about 10 BPM every three days, an increase of 100 BPM in the first month. After peaking at about 9.2 weeks after the LMP, it decelerates to about 150 BPM (+/-25 BPM) during the 15th week after the LMP. After the 15th week the deceleration slows reaching an average rate of about 145 (+/-25 BPM) BPM at term. Imaging In the first trimester, heartbeat can be visualized and the heart rate quantified by obstetric ultrasonography. A study of 32 normal pregnancies came to the result a fetal heartbeat was visible at a mean human chorionic gonadotropin (hCG) level of 10,000 UI/l (range 8650-12,200). Obstetric ultrasonography can also use doppler technique on key vessels such as the umbilical artery can detect abnormal flow. In later stages of pregnancy, a simple Doppler fetal monitor can quantify the fetal heart rate. During childbirth, the parameter is part of cardiotocography, which is where the fetal heartbeat and uterine contractions are continuously recorded. Heart rates Starting at week 5 the embryonic heart rate accelerates by 3.3 bpm per day for the next month. Before this, the embryo possesses a tubular heart. The embryonic heart begins to beat at approximately the same rate as the mother's, which is typically 80 to 85 bpm. The approximate fetal heart rate for weeks 5 to 9 (assuming a starting rate of 80): Week 5 starts at 80 and ends at 103 bpm Week 6 starts at 103 and ends at 126 bpm Week 7 starts at 126 </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> and ends at 149 bpm Week 8 starts at 149 and ends at 172 bpm At week 9 the embryonic heart tends to beat within a range of 155 to 195 bpm. By the end of week 9, the embryonic heart has developed septa and valves, and has all four chambers. At this point, the fetal heart rate begins to decrease, and generally falls within the range of 120 to 160 bpm by week 12. Additional images References Category:Embryology of cardiovascular system </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Mifflintown Formation The Mifflintown Formation is a geologic formation in Maryland. It preserves fossils dating back to the Silurian period. See also List of fossiliferous stratigraphic units in Maryland Paleontology in Maryland References Category:Silurian Maryland Category:Silurian geology of Pennsylvania Category:Silurian southern paleotemperate deposits </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Frank Rushton (athlete) Frank Rushton (born 19 April 1909, date of death unknown) was a South African sprinter. He competed in the men's 4 × 400 metres relay at the 1936 Summer Olympics. References Category:1909 births Category:Year of death missing Category:Athletes (track and field) at the 1936 Summer Olympics Category:South African male sprinters Category:South African male hurdlers Category:Olympic athletes of South Africa Category:Place of birth missing </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> 1965 in Turkey Events in the year 1965 in Turkey. Parliament 12th Parliament of Turkey (up to 10 October) 13th Parliament of Turkey Incumbents President – Cemal Gürsel Prime Minister İsmet İnönü (up to 20 February) Suat Hayri Ürgüplü (20 February-27 October) Süleyman Demirel (from 27 October) Leader of the opposition Süleyman Demirel (up to 20 February) İsmet İnönü (from 20 February) Ruling party and the main opposition Ruling party Republican People’s Party (CHP) (up to 20 February) A coalition of Justice Party (AP), New Turkey Party (YTP) and Republican Villagers Nation Party (CKMP) and Nation Party (MP) (20 February-27 October) Justice Party (AP) (from 27 October) Main opposition Justice Party (AP) (up to 20 February) Republican People’s Party (CHP) (from 20 February) Cabinet 28th government of Turkey (up to 20 February) 29th government of Turkey (20 February-27 October) 30th government of Turkey (from 27 October) Events 13 February – Annual budget of the 28th government of Turkey was rejected. (End of İsmet İnönü government and a shorth term caretaker government)) 19 March – Gas explosion in Yeniçeltek mine, Amasya Province. 69 deaths 13 June – Fenerbahçe won the championship of Turkish football league. 1 July - Road Water Electricity General Directorate was founded. 10 October – General elections . AP 240 seats, CHP 134 seats, MP 31 seats, YTP 19 seats, TİP 14 seats, CKMP 11 seats, Indep 1 seat. (TİP was the first marxist party to gain seats in Turkish parliament) 24 October – Census Population 31351421. 18 December – Turkey declared that Zürich and London Agreement concerning the Cyprus issue was still in effect Births 24 February – Tülay Keçialan (Asya), singer 5 April – Aykut Kocaman footballer 18 August – Hayrünnisa Gül, ex President Abdullah Gül's wife 12 September – Seden Gürel, singer Deaths 18 July – Refik Halit Karay (aged 77), novelist 15 August – Zihni Derin (aged 85), agronomis Gallery See also 1964–65 1.Lig References Category:Years of the 20th century in Turkey </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Grandmothers for Peace Grandmothers for Peace (a.k.a. Grandmothers for Peace International) is an organization started by Barbara Wiedner (1928-2001) in May 1982 in Sacramento, California in the USA, after the mass media became "captivated by the image of a grandmother Barbara Wiedner risking jail through non-violent civil disobedience in an effort to save the planet from nuclear annihilation". The organisation includes over 40 chapters around the USA and chapters in Berlin, Germany, Romania, South Africa and the United Kingdom. In 2006, the group participated in the Don't Attack Iran Coalition which opposed a military attack against Iran. References Category:Peace organizations based in the United States Category:Organizations established in 1982 Category:1982 establishments in California </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Cilada.com Cilada.com is a 2011 Brazilian comedy film directed by José Alvarenga Jr., released on 8 July 2011 in Brazil. It's inspired by the sitcom Cilada. Cast Bruno Mazzeo as Bruno Fernanda Paes Leme as Fernanda Augusto Madeira as Sandro Carol Castro as Mônica Fabiula Nascimento as Suzy Fúlvio Stefanini as Dr. Leoni Sérgio Loroza as Marco 'Marconha' André (as Serjão Loroza) Thelmo Fernandes as Gerson Dani Calabresa as Regina Kelly References External links Category:2011 films Category:2010s comedy films Category:Brazilian films Category:Brazilian comedy films Category:Portuguese-language films Category:Films shot in Rio de Janeiro (city) </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Jan Callebaut Jan Callebaut (Geraardsbergen, 8 November 1955) is a Belgian communication and marketing advisor and entrepreneur. Education Jan holds an MA in Marketing and Distribution and an MA in Diplomatics of the University of Ghent. As an academic researcher he developed psychoanalytical market research techniques to consumer behavior by applying the views of Alfred Adler. Career In 1987 Jan co-founded Censydiam (Center for Systematic Diagnostics in Marketing) now Ipsos Censydiam. He supplied qualitative marketing research and business strategies for companies like Coca-Cola, Heineken, Unilever, Volvo, InBev, Johnson & Johnson, Friesland and was engaged in the public sector with projects for Foster Parents Plan and Amnesty International. Jan Callebaut also introduced the brand approach into the VRT, the Flemish Radio and Television Network, Belgium’s leading broadcasting and media concern. In 2007 Jan founded Callebaut&Co and became CEO of WHy5Research, an international diagnostic market research agency. More recently Jan co-founded Callebaut Collective, an ecosystem of strategic advisors and partners in execution, where he currently acts as managing partner. Vision Jan is one of the founders of the modern diagnostic market research. His insights on consumer behaviour based on a psychoanalytic approach is adopted by various marketing experts like Philip Kotler and Kevin Lane Keller (Marketing Management) and known as the censydiam model . The model is developed to understand better the mechanisms of human motivation. Personal life Jan Callebaut is married to Gisèle Keppens with whom he has two children, Anouk (1994) and Kilian (1998). Bibliography The Naked Consumer (1994, Censydiam Institute) Never too late to grow old (1996, Censydiam Institute) Nooit te laat om oud te worden (1996, Censydiam Institute) Cross-Culturally Correct Marketing (1997, Garant) Motivational Marketing Research Revisited (1999, (Garant Uitgevers) Deux ou trois choses que je sais de la France (1999) – Lannoo ) Cross-Cultural Window on Consumer Behaviour (2000, Garant ) El Nuevo Modelo Diagnóstico Para El Márketing (2000, Garant ) Understanding Chinese consumers (2000, Garant ) Why yesterday tells of tomorrow (2001, Garant ) Warum Gestern über Morgen erzählt (2001, Garant ) The Naked Consumer Today (2002, Garant ) Marketing is menselijk : collection of columns for the Belgian newspaper De Standaard (2004, Garant ) Colour and Emotion: humanising the message (2005, Synovate) Consumentengeheimen (2009, Davidsfonds) Why you do what you do (2011, Why5 Research ) External links Why5Research Callebaut&Co Callebaut Collective Category:1955 births Category:Living people Category:Belgian businesspeople </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Billie Jean King Billie Jean King (née Moffitt; born November 22, 1943) is an American former World No. 1 professional tennis player. King won 39 Grand Slam titles: 12 in singles, 16 in women's doubles, and 11 in mixed doubles. She won the singles title at the inaugural WTA Tour Championships. She often represented the United States in the Federation Cup and the Wightman Cup. She was a member of the victorious United States team in seven Federation Cups and nine Wightman Cups. For three years, she was the United States' captain in the Federation Cup. King is an advocate for gender equality and has long been a pioneer for equality and social justice. In 1973, at age 29, she won the "Battle of the Sexes" tennis match against the 55-year-old Bobby Riggs. She was also the founder of the Women's Tennis Association and the Women's Sports Foundation. She was also instrumental in persuading cigarette brand Virginia Slims to sponsor women's tennis in the 1970s and went on to serve on the board of their parent company Philip Morris in the 2000s. Regarded by many in the sport as one of the greatest women's tennis players of all time, King was inducted into the International Tennis Hall of Fame in 1987. The Fed Cup Award of Excellence was bestowed on her in 2010. In 1972, she was the joint winner, with John Wooden, of the Sports Illustrated Sportsman of the Year award and was one of the Time Persons of the Year in 1975. She has also received the Presidential Medal of Freedom and the Sunday Times Sportswoman of the Year lifetime achievement award. She was inducted into the National Women's Hall of Fame in 1990, and in 2006, the USTA National Tennis Center in New York City was renamed the USTA Billie Jean King National Tennis Center. In 2018, she won the BBC Sports Personality of the Year Lifetime Achievement Award. Early life Billie Jean Moffitt was born in Long Beach, California, into a conservative Methodist family, the daughter of Betty (née Jerman), a housewife, and Bill Moffitt, a firefighter. Her family was athletic; her mother excelled at swimming, her father played basketball, baseball and ran track. Her younger brother, Randy Moffitt, became a Major League Baseball pitcher, pitching for 12 years in the major leagues for the San Francisco Giants, Houston Astros, and Toronto Blue Jays. She also excelled at baseball and softball as a child, playing shortstop at 10 years old on a team with girls 4–5 years older than her. The team went on to win the Long Beach softball championship. She switched from softball to tennis at age 11, because her parents suggested she should find a more ladylike sport. She saved her own money, $8, to buy her first racket. She learned tennis on the many free public courts in Long Beach, taking advantage of the free lessons tennis professional Clyde Walker offered at those courts. One of the city's tennis facilities has subsequently been named the Billie Jean Moffitt King Tennis Center. As a kid playing in her first tennis tournaments, she was often hindered by her aggressive playing style. Bob Martin, sportswriter for the Long Beach, Press-Telegram wrote about her success in a weekly tennis column. King's church minister was the 1952 and 1956 pole-vaulting champion, the Rev. Bob Richards. One day, when King was 13 or 14, Richards asked her, "What are you going to do with your life?" She said: "Reverend, I'm going to be the best tennis player in the world." She attended Long Beach Polytechnic High School. After </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> graduating in 1961, she attended Los Angeles State College now California State University, Los Angeles (Cal State LA). She did not graduate, leaving school in 1964 to focus on tennis. While attending Cal State, she met Larry King in a library in 1963. The pair became engaged while still in school when Billie Jean was 20 and Larry 19 years old and married on September 17, 1965 in Long Beach. Career King's triumph at the French Open in 1972 made her only the fifth woman in tennis history to win the singles titles at all four Grand Slam events, a "career Grand Slam." She also won a career Grand Slam in mixed doubles. In women's doubles, only the Australian Open eluded her. King won a record 20 career titles at Wimbledon – six in singles, 10 in women's doubles, and four in mixed doubles. King played 51 Grand Slam singles events from 1959 through 1983, reaching at least the semifinals in 27 and at least the quarterfinals in 40 of her attempts. King was the runner-up in six Grand Slam singles events. An indicator of her mental toughness in Grand Slam singles tournaments was her 11–2 career record in deuce third sets, i.e., third sets that were tied 5–5 before being resolved. King won 129 singles titles, 78 of which were WTA titles, and her career prize money totaled US$1,966,487. In Federation Cup finals, she was on the winning United States team seven times, in 1963, 1966, 1967, and 1976 through 1979. Her career win–loss record was 52–4. She won the last 30 matches she played, including 15 straight wins in both singles and doubles. In Wightman Cup competition, her career win–loss record was 22–4, winning her last nine matches. The United States won the cup ten of the 11 years that she participated. In singles, King was 6–1 against Ann Haydon-Jones, 4–0 against Virginia Wade, and 1–1 against Christine Truman Janes. The early years: 1959 through 1963 As King began competing in 1959, she began working with new coaches including Frank Brennan and Alice Marble, who had won 18 Grand Slam titles as a player herself. She made her Grand Slam debut at the 1959 U.S. Championships at age 15. She lost in the first round. She began playing at local, regional, and international tennis championships. Sports Illustrated already claimed her as "one of the most promising youngsters on the West Coast." She won her first tournament the next year in Philadelphia at the 1960 Philadelphia and District Grass Court Championships. At her second attempt at the U.S. Championships, King made it to the third round, losing to Bernice Carr Vukovich of South Africa. Also in 1960, she reached the final of the National Girl's 18 and Under Championships, losing to Karen Hantze Susmen. Her national tennis ranking improved from number 19 in 1959 to number 4 1960. Despite the success, Marble terminated her professional relationship with King because for reasons stemming from King's ambition. King first gained international recognition in 1961 when the Long Beach Tennis Patrons, Century Club, and Harold Guiver raised $2,000 to send her to Wimbledon. There, she won the women's doubles title in her first attempt while partnering Karen Hantze. King was 17 and Hantze was 18, making them the youngest team to win the Wimbledon Doubles Title. King had less luck that year in the 1961 Wimbledon Women's Singles, losing to fifth-seeded Yola Ramírez Ochoa in a two-day match on Centre Court. For the 1962 singles tournament at Wimbledon, King upset Margaret Court, the World No. 1 and top seed, in a second </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> round match by attacking Court's forehand This was the first time in Wimbledon history that the women's top seed had lost her first match. That same year, King and Hantze repeated their doubles victory at Wimbledon. In 1963, King again faced Margaret Court at Wimbledon. This time they met in the finals with Court prevailing. 1964 In 1964, King won four relatively minor titles but lost to Margaret Court in the Wimbledon semifinals. She defeated Ann Haydon-Jones at both the Wightman Cup and Fed Cup but lost to Court in the final of the Federation Cup. At the U.S. Championships, fifth-seeded Nancy Richey Gunter upset third-seeded King in the quarterfinals. Late in the year, King decided to make a full-time commitment to tennis. While a history major at Los Angeles State College King made the decision to play full-time when businessman Robert Mitchell, offered to pay her way to Australia so that she could train under the great Australian coach Mervyn Rose. While in Australia, King played three tournaments that year and lost in the quarterfinals of the Queensland Grass Court Championships, the final of the New South Wales Championships (to Court), and the third round of the Victorian Championships. 1965 In early 1965, King continued her three-month tour of Australia. She lost in the final of the South Australian Championships and the first round of the Western Australia Championships. At the Fed Cup in Melbourne, she defeated Ann Haydon-Jones to help the United States defeat the United Kingdom in the second round. However, Margaret Court again defeated her in the final. At the Australian Championships two weeks later, she lost to Court in the semifinals in two sets. At Wimbledon, she again lost in the semifinals, this time in three sets to Maria Bueno. Her last tournament of the year was the U.S. Championships, where she defeated Jones in the quarterfinals and Bueno in the semifinals. In the final, King led 5–3 in both sets, was two points from winning the first set, and had two set points in the second set before losing to Court in straight sets. She said that losing while being so close to winning was devastating, but the match proved to her that she was "good enough to be the best in the world. I'm going to win Wimbledon next year." She won six tournaments during the year. For the first time in 81 years, the annual convention of the United States Lawn Tennis Association overruled its ranking committee's recommendation to award her the sole U.S. No. 1 position and voted 59,810 to 40,966 to rank Nancy Richey Gunter and King as co-U.S. No. 1. Prime competitive years: 1966 through 1975 Overview of these years Six of King's Grand Slam singles titles were at Wimbledon, four were at the U.S. Championships/Open, one was at the French Open, and one was at the Australian Championships. King reached the final of a Grand Slam singles tournament in 16 out of 25 attempts and had a 12–4 win–loss record in those finals. In the nine tournaments that she failed to reach the final, she was a losing semifinalist twice and a losing quarter finalist five times. From 1971 through 1975, she won seven of the ten Grand Slam singles tournaments she played. She won the last seven Grand Slam singles finals she contested, six of them in straight sets and four of them against Evonne Goolagong. All but one of her Grand Slam singles titles were on grass. King's Grand Slam record from 1966 through 1975 was comparable to that of Margaret Court, her primary rival during </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> these years. One or both of these women played 35 of the 40 Grand Slam singles tournaments held during this period, and together they won 24 of them. During this period, Court won 31 of her career 64 Grand Slam titles, including 12 of her 24 Grand Slam singles titles, 11 of her 19 Grand Slam women's doubles titles, and eight of her 21 Grand Slam mixed doubles titles. Court reached the final of a Grand Slam singles tournament in 14 out of 25 attempts and had a 12–2 win–loss record in those finals. Court won seven of the 12 Grand Slam finals she played against King during these years, including 2–1 in singles finals, 4–1 in women's doubles finals, and 1–3 in mixed doubles finals. King was the year-ending World No. 1 in six of the ten years from 1966 through 1975. She was the year-ending World No. 2 in three of those years and the World No. 3 in the other year. King won 97 of her career 129 singles titles during this period and was the runner-up in 36 other tournaments. 1966 In 1966, King defeated Dorothy "Dodo" Cheney (then 49 years old) for the first time in five career matches, winning their semifinal at the Southern California Championships 6–0, 6–3. King also ended her nine match losing streak to Margaret Court by defeating her in the final of the South African Tennis Championships. At the Wightman Cup just before Wimbledon, King defeated Virginia Wade and Ann Haydon-Jones. After thirteen unsuccessful attempts to win a Grand Slam singles title from 1959 through 1965, King at the age of 22 finally won the first of her six singles titles at Wimbledon and the first of twelve Grand Slam singles titles overall, defeating Court in the semifinals 6–3, 6–3 and Maria Bueno in the final. King credited her semifinal victory to her forehand down the line, a new shot in her repertoire. She also said that the strategy for playing Court is, "Simple. Just chip the ball back at her feet." At the U.S. Championships, an ill King was upset by Kerry Melville Reid in the second round. 1967 King successfully defended her title at the South African Tennis Championships in 1967, defeating Maria Bueno in the final. She played the French Championships for the first time in her career, falling in the quarterfinals to Annette Van Zyl DuPlooy of South Africa. At the Federation Cup one week later in West Germany on clay, King won all four of her matches, including victories over DuPlooy, Ann Haydon-Jones, and Helga Niessen Masthoff. King then successfully switched surfaces and won her second consecutive Wimbledon singles title, defeating Virginia Wade in the quarterfinals 7–5, 6–2 and Jones. At the Wightman Cup, King again defeated Wade and Jones. King won her second Grand Slam singles title of the year when she won the U.S. Championships for the first time and without losing a set, defeating Wade, DuPlooy, Françoise Dürr, and Jones in consecutive matches. Jones pulled her left hamstring muscle early in the final and saved four match points in the second set before King prevailed. King won the singles, women's doubles, and mixed doubles titles at both Wimbledon and the U.S. Championships, the first woman to do that since Alice Marble in 1939. King then returned to the Australian summer tour in December for the first time since 1965, playing seven events there and Judy Tegart-Dalton in six of those events (winning four of their matches). King lost in the quarterfinals of the New South Wales Championships in Sydney to Dalton after </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> King injured her left knee in the second game of the third set of that match. However, King won the Victorian Championships in Melbourne the following week, defeating Dalton, Reid, and Lesley Turner Bowrey in the last three rounds. At a team event in Adelaide, King won all three of her singles and doubles matches to help the U.S. defeat Australia 5–1. To finish the year, King lost to Dalton in the final of the South Australian Championships in Adelaide. 1968 In early 1968, King won three consecutive tournaments to end her Australian tour. In Perth, King won the Western Australia Championships, defeating Margaret Court in the final. In Hobart, King won the Tasmanian Championships by defeating Judy Tegart-Dalton in the final. King then won the Australian Championships for the first time, defeating Dalton in the semifinals and Court in the final. King continued to win tournaments upon her return to the United States, winning three indoor tournaments before Nancy Richey Gunter defeated King in the semifinals of the Madison Square Garden Challenge Trophy amateur tournament in New York City before 10,233 spectators. The match started with Gunter taking a 4–2 lead in the first set, before King won 9 of the next 10 games. King served for the match at 5–1 and had a match point at 5–3 in the second set; however, she lost the final 12 games and the match 4–6, 7–5, 6–0. King then won three consecutive tournaments in Europe before losing to Ann Haydon-Jones in the final of a professional tournament at Madison Square Garden. Playing the French Open for only the second time in her career and attempting to win four consecutive Grand Slam singles titles (a "non-calendar year Grand Slam"), King defeated Maria Bueno in a quarterfinal before losing to Gunter in a semifinal 2–6, 6–3, 6–4. King rebounded to win her third consecutive Wimbledon singles title, defeating Jones in the semifinals and Dalton in the final. At the US Open, King defeated Bueno in a semifinal before being upset in the final by Virginia Wade. On September 24, she had surgery to repair cartilage in her left knee and did not play in tournaments the remainder of the year. King said that it took eight months (May 1969) for her knee to recover completely from the surgery. In 1977, King said that her doctors predicted in 1968 that her left knee would allow her to play competitive tennis for only two more years. 1969 King participated in the 1969 Australian summer tour for the second consecutive year. Unlike the previous year, King did not win a tournament. She lost in the quarterfinals of the Tasmanian Championships and the semifinals of the New South Wales Championships. At the Australian Open, King defeated 17-year-old Evonne Goolagong in the second round 6–3, 6–1 and Ann Haydon-Jones in a three-set semifinal before losing to Margaret Court in a straight-sets final. The following week, King lost in the semifinals of the New Zealand Championships. Upon her return to the United States, King won the Pacific Coast Pro and the Los Angeles Pro. King then won two tournaments in South Africa, including the South African Open. During the European summer clay court season, King lost in the quarterfinals of both the Italian Open and the French Open. On grass at the Wills Open in Bristol, United Kingdom, King defeated Virginia Wade in the semifinals (6–8, 11–9, 6–2) before losing to Court. At Wimbledon, King lost only 13 points while defeating Rosemary Casals in the semifinals 6–1, 6–0; however, Jones upset King in the final and prevented King from </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> winning her fourth consecutive singles title there. The week after, King again defeated Wade to win the Irish Open for the second time in her career. In the final Grand Slam tournament of the year, King lost in the quarterfinals of the US Open to Nancy Richey Gunter 6–4, 8–6. This was the first year since 1965 that King did not win at least one Grand Slam singles title. King finished the year with titles at the Pacific Southwest Open in Los Angeles, the Stockholm Indoors, and the Midland (Texas) Pro. She said during the Pacific Southwest Open, "It has been a bad year for me. My left knee has been OK, but I have been bothered by a severe tennis elbow for seven months. I expect to have a real big year in 1970, though, because I really have the motivation now. I feel like a kid again." 1970 In 1970, Margaret Court won all four Grand Slam singles tournaments and was clearly the World No. 1. King lost to Court three times in the first four months of the year, in Philadelphia, Dallas, and Johannesburg (at the South African Open). Court, however, was not totally dominant during this period as King defeated her in Sydney and Durban, South Africa. Where Court dominated was at the Grand Slam tournaments. King did not play the Australian Open. King had leg cramps and lost to Helga Niessen Masthoff of West Germany in the quarterfinals of the French Open 2–6, 8–6, 6–1. At Wimbledon, Court needed seven match points to defeat King in the final 14–12, 11–9 in one of the greatest women's finals in the history of the tournament. On July 22, King had right knee surgery, which forced her to miss the US Open. King returned to the tour in September, where she had a first round loss at the Virginia Slims Invitational in Houston and a semifinal loss at the Pacific Coast Championships in Berkeley, California. To close out the year, King in November won the Virginia Slims Invitational in Richmond, Virginia and the Embassy Indoor Tennis Championships in London. During the European clay court season, King warmed-up for the French Open by playing in Monte Carlo (losing in the semifinals), winning the Italian Open (saving three match points against Virginia Wade in the semifinals), playing in Bournemouth (losing to Wade in the quarterfinals), and playing in Berlin (losing to Masthoff in the semifinals). The Italian Open victory was the first important clay court title of King's career. Along the way, she defeated Masthoff in a three-set quarterfinal and Wade in a three-set semifinal, saving two match points at 4–5 in the second set. The twelfth game of that set (with King leading 6–5) had 21 deuces and lasted 22 minutes, with Wade saving seven set points and holding sixteen game points before King won. In Wightman Cup competition two weeks before Wimbledon but played at the All England Club, King defeated both Wade and Ann Haydon-Jones in straight sets. 1971 Although King won only one Grand Slam singles title in 1971, this was the best year of her career in terms of tournaments won (17). According to the International Tennis Hall of Fame, she played in 31 singles tournaments and compiled a 112–13 win–loss record. She started the year by winning eight of the first thirteen tournaments she played, defeating Rosemary Casals in seven finals. King's five losses during this period were to Françoise Dürr (twice), Casals (once), Ann Haydon-Jones (once), and Chris Evert (in St. Petersburg). At the time, King said that retiring from the match with </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Evert after splitting the first two sets was necessary because of leg cramps. But in early 1972, King admitted that cramps associated with the abortion caused the retirement. At the tournament in Hurlingham, United Kingdom in early May, King lost a second round match to an old rival, Christine Truman Janes (now 30 years old), 6–4, 6–2. But King recovered the next week to win the German Open in Hamburg on clay. Four weeks later at the Queen's Club tournament in London, King played Margaret Court for the first time in 1971, losing their final. At Wimbledon, King defeated Janes in the fourth round (6–2, 7–5) and Durr in the quarterfinals before losing unexpectedly to Evonne Goolagong in the semifinals 6–4, 6–4. Two weeks after Wimbledon, King won the grass court tournament in Hoylake, United Kingdom, beating Virginia Wade, Court, and Casals in the last three rounds. She then played two clay court tournaments in Europe, winning neither, before resuming play in the United States. In August, King won the indoor Houston tournament and the U.S. Women's Clay Court Championships in Indianapolis. King then switched back to grass and won the US Open without losing a set, defeating Evert in the semifinals (6–3, 6–2) and Casals in the final. King then won the tournaments in Louisville, Phoenix, and London (Wembley Pro). King and Casals both defaulted at 6–6 in the final of the Pepsi Pacific Southwest Open in Los Angeles in September when their request to remove a lineswoman was denied, eventually resulting in the United States Lawn Tennis Association fining both players US$2,500. To end the year, King played two tournaments in New Zealand but did not win either. She lost in Christchurch to Durr and in Auckland to Kerry Melville Reid. 1972 King won three Grand Slam singles titles in 1972, electing not to play the Australian Open despite being nearby when she played in New Zealand in late 1971. King said, "I was twenty-eight years old, and I was at the height of my powers. I'm quite sure I could have won the Grand Slam [in] ... 1972, but the Australian was such a minor-league tournament at that time.... More important, I did not want to miss any Virginia Slims winter tournaments. I was playing enough as it was." Her dominance was aided by rival Margaret Court's absence from the tour due to childbirth during most of the 1972 season. At the beginning of the year, King failed to win eight of the first ten tournaments she played. She won the title in San Francisco in mid-January. But then King lost in Long Beach to Françoise Dürr (although King claimed in her 1982 autobiography that she intentionally lost the match because of an argument with her husband) and in Fort Lauderdale on clay to Chris Evert 6–1, 6–0. The inconsistent results continued through mid-April, in Oklahoma City (losing in the quarterfinals); Washington, D.C. (losing in the second round); and Dallas (losing to Nancy Richey Gunter after defeating Evert in the quarterfinals 6–7(4–5), 6–3, 7–5 and Evonne Goolagong in the semifinals 1–6, 6–4, 6–1). King won the title in Richmond; however, one week later, King lost in the semifinals of the tournament in San Juan. This was followed in successive weeks by a loss in the Jacksonville final to Marie Neumannova Pinterova and in a St. Petersburg semifinal to Evert (6–2, 6–3). King did not lose again until mid-August, winning six consecutive tournaments. She won the tournaments in Tucson and Indianapolis. King then won the French Open without losing a set and completed a career Grand Slam. She </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> defeated Virginia Wade in the quarterfinals, Helga Niessen Masthoff in the semifinals, and Goolagong in the final. On grass, King then won the Wimbledon warm-up tournaments in Nottingham and Bristol and won Wimbledon itself for the fourth time. She lost only one set during the tournament, to Wade in the quarterfinals. That was followed by straight set wins over Rosemary Casals and Goolagong. When the tour returned to the United States, King did not win any of the three tournaments she played before the US Open, including a straight sets loss to Margaret Court in Newport. At the US Open, however, King won the tournament without losing a set, including a quarterfinal win over Wade, a semifinal defeat of Court, and a final win over Kerry Melville Reid. King finished the year by winning the tournaments in Charlotte and Phoenix (defeating Court in the final of both), a runner-up finish in Oakland (losing to Court), and a semifinal finish in Boca Raton (losing to Evert). 1973 1973 was Margaret Court's turn to win three Grand Slam singles titles, failing to win only Wimbledon, and was the clear World No. 1 for the year; this was her first full season since winning the Grand Slam in 1970, as she had missed significant portions of 1971 and 1972 due to childbirth. As during the previous year, King started 1973 inconsistently. She missed the first three Virginia Slims tournaments in January because of a wrist injury. She then lost in the third round at the Virginia Slims of Miami tournament but won the Virginia Slims of Indianapolis tournament, defeating Court in the semifinals 6–7, 7–6, 6–3 and Rosemary Casals in the final. The semifinal victory ended Court's 12-tournament and 59-match winning streaks, with King saving at least three match points when down 5–4 (40–0) in the second set. Indianapolis was followed by five tournaments that King failed to win (Detroit, Boston, Chicago, Jacksonville, and the inaugural Family Circle Cup in Hilton Head, South Carolina). King lost to Court in two of those tournaments. After deciding not to defend her French Open singles title, King won four consecutive tournaments, including her fifth Wimbledon singles title when she defeated Kerry Melville Reid in the quarterfinals, Evonne Goolagong in the semifinals on her eighth match point, and Chris Evert in the final. King lost only nine points in the 6–0 bageling of Evert in the first set of their final. King also completed the Triple Crown at Wimbledon (winning the singles, women's doubles, and mixed doubles titles in the same year), thus becoming the first, and only, player to do so at Wimbledon in the Open Era. In none of the preceding tournaments, however, did King play Court. Their rivalry resumed in the final of the Virginia Slims of Nashville tournament, where Court won for the third time in four matches against King in 1973. (This was the last ever singles match between those players, with Court winning 21 and King 13 of their 34 matches.) Three weeks later at the US Open, King retired from her fourth-round match with Julie Heldman while ill and suffering from the oppressive heat and humidity. When Heldman complained to the match umpire that King was taking too long between games, King reportedly told Heldman, "If you want the match that badly, you can have it!" The Battle of the Sexes match against Bobby Riggs was held in the middle of the Virginia Slims of Houston tournament. King won her first and second round matches three days before playing Riggs, defeated Riggs, won her quarterfinal match the day after the </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Riggs match, and then lost the following day to Casals in the semifinals 7–6, 6–1. According to King, "I had nothing left to give." To end the year, King won tournaments in Phoenix, Hawaii, and Tokyo and was the runner-up in Baltimore. Battle of the Sexes In 1973, King defeated Bobby Riggs in an exhibition match, winning $100,000. Riggs had been a top men's player in the 1930s and 1940s in both the amateur and professional ranks. He won the Wimbledon men's singles title in 1939, and was considered the World No. 1 male tennis player for 1941, 1946, and 1947. He then became a self-described tennis "hustler" who played in promotional challenge matches. Claiming that the women's game was so inferior to the men's game that even a 55-year-old like himself could beat the current top female players, he challenged and defeated Margaret Court 6–2, 6–1. King, who previously had rejected challenges from Riggs, then accepted a lucrative financial offer to play him for $100,000, winner-take-all. Dubbed "the Battle of the Sexes", the Riggs-King match took place at the Houston Astrodome in Texas on September 20, 1973. The match garnered huge publicity. In front of 30,492 spectators and a television audience estimated at 50 million people (U.S.), and 90 million in 37 countries, 29-year-old King beat the 55-year-old Riggs 6–4, 6–3, 6–3. The match is considered a very significant event in developing greater recognition and respect for women's tennis. King said, "I thought it would set us back 50 years if I didn't win that match. It would ruin the women's [tennis] tour and affect all women's self-esteem," and that "To beat a 55-year-old guy was no thrill for me. The thrill was exposing a lot of new people to tennis." 1974 King won five of the first seven tournaments she contested in 1974. She won the Virginia Slims of San Francisco, defeating Nancy Richey Gunter in the semifinals and Chris Evert in the final. The following week in Indian Wells, California, King again defeated Gunter in the semifinals but lost to Evert in the final. King then won tournaments in Fairfax, Virginia and Detroit before losing a semifinal match to Virginia Wade in Chicago. King won both tournaments she played in March, defeating Gunter in the Akron, Ohio final and Evert at the U.S. Indoor Championships final. Olga Morozova then upset King in her next two tournaments, at Philadelphia in the final and at Wimbledon in a quarterfinal 7–5, 6–2. Afterward, King did not play a tour match until the US Open, where she won her fourth singles title and third in the last four years. She defeated Rosemary Casals in a straight sets quarterfinal, avenged in the semifinals her previous year's loss to Julie Heldman, and narrowly defeated Evonne Goolagong in the final. King did not reach a tournament final during the remainder of the year, losing to Heldman in an Orlando semifinal, Wade in a Phoenix semifinal, and Goolagong in a semifinal of the tour-ending Virginia Slims Championships in Los Angeles. 1975 In 1975, King played singles only half the year, as she retired (temporarily, as it turned out) from tournament singles competition immediately after winning her sixth Wimbledon singles title. She began the year in San Francisco, defeating Françoise Dürr and Virginia Wade before losing to Chris Evert in the final. The following week, King won the Sarasota, Florida tournament, defeating Evert in the final 6–3, 6–2. Evert said immediately after the final, which was her thirteenth career match with King, "I think that's the best that Billie Jean has ever played. I hit some </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> great shots but they just kept coming back at me." Looking back at that match, King said, "I probably played so well because I had to, for the money. Out of frustration comes creativity. Right?" Two months later, Wade defeated King in the semifinals of the Philadelphia tournament. At the Austin, Texas tournament in April, King defeated Evonne Goolagong 6–1, 6–3 before losing to Evert in the final. As King was serving for the match at 6–5 in the third set, a disputed line call went in Evert's favor. King said after the match that she was cheated out of the match and that she had never been angrier about a match. King played only one of the Wimbledon warm-up tournaments, defeating Olga Morozova in the Eastbourne semifinals before losing to Wade in the final. Seeded third at Wimbledon, King defeated seventh seeded Morozova in the quarterfinals (6–3, 6–3) and then top seeded Evert in the semifinals (2–6, 6–2, 6–3) after being down 3–0 (40–15) in the final set. Evert blamed her semifinal defeat on a loss of concentration when she saw Jimmy Connors, her former fiancé, escorting Susan George into Centre Court. King, however, believes that the match turned around because King planned for and totally prepared for Wimbledon that year and told herself when she was on the verge of defeat, "Hey, Billie Jean, this is ridiculous. You paid the price. For once, you looked ahead. You're supposed to win. Get your bahoola in gear." King then defeated fourth seeded Goolagong Cawley in the second most lopsided women's final ever at Wimbledon (6–0, 6–1). King called her performance a "near perfect match" and said to the news media, "I'm never coming back." The later years: 1976 through 1990 1976 Except for five Federation Cup singles matches that she won in straight sets in August, King played only in doubles and mixed doubles events from January through September. She partnered Phil Dent to the mixed doubles title at the US Open. She lost to Dianne Fromholtz Balestrat in both of the singles tournaments she played the remainder of the year. Looking back, King said, "I wasted 1976. After watching Chris Evert and Evonne [Goolagong] Cawley play the final at Wimbledon I asked myself what I was doing. So, despite my age and the operations, the Old Lady came back...." King had knee surgery for the third time on November 9, this time on her right knee, and did not play the remainder of the year. 1977 King spent the first three months of the year rehabilitating her right knee after surgery in November 1976. In March 1977, King requested that the Women's Tennis Association (WTA) exercise its right to grant a wild card entry to King for the eight-player Virginia Slims Championships at Madison Square Garden in New York City. Margaret Court, who finished in sixth place on the Virginia Slims points list, failed to qualify for the tournament because she did not play enough Virginia Slims tournaments leading up to the championships. This left a spot open in the draw, which the WTA filled with Mima Jaušovec. King then decided to play the Lionel Cup tournament in San Antonio, Texas, which the WTA harshly criticized because tournament officials there had allowed transsexual Renée Richards to enter. Chris Evert, Martina Navratilova, and Betty Stöve (president of the WTA) criticized King's decision because of Richards's unresolved and highly controversial status on the women's tennis tour. Evert said she was disappointed with King and that until Richards's status was resolved, "all of the women should stick together." Navratilova said, "Billie Jean </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> is a bad girl pouting. She made a bad decision. She's mad because she could not get what she wanted." Stöve said that if King had wanted the competition, "[T]here are plenty of men around here she could've played with. She didn't have to choose a 'disputed' tournament." The draw in San Antonio called for King to play Richards in the semifinals had form held; however, Richards lost in the quarterfinals. King eventually won the tournament. At the clay court Family Circle Cup in late March, King played for the last time her long-time rival Nancy Richey Gunter in the first round. King won 0–6, 7–6, 6–2. She defeated another clay court specialist, Virginia Ruzici, in the second round before winning only one game from Evert in the final. At Wimbledon in the third round, King played Maria Bueno for the last time, with King winning 6–2, 7–5. In the quarterfinals, Evert defeated King for the first time at a Grand Slam singles tournament and for the first time on grass 6–1, 6–2 in just 46 minutes. Evert said it was the best match she had ever played on grass up to that point in her career, and King said, "No excuses. Let's forget knees, ankles, toes, everything else. She just played beautiful tennis. I don't think many players would've beaten her today." King also said after the match, "Maybe I can be happy being number eight instead of number one. At this stage, just playing, that's winning enough for me." But when asked about retirement, King said, "Retire? Quit tournament tennis? You gotta be kidding. It just means I've got a lot more work. I've got to make myself match tough ... mentally as well as physically. I gotta go out and kill myself for the next six months. It's a long, arduous process. I will suffer. But I will be back." There was a small historic note at Wimbledon 1977 in that it was the first time ever that King competed at the championships that she did not reach a final. From her debut in 1961 until 1976, she had played in the final of one of the three championship events for women every year. Perhaps there was irony in this in that as the Wimbledon champion with the most titles in its history, the event was celebrating its centenary in the year King failed to make a final for the first time. The only other years she competed at the championship and did not feature in a final were 1980 and 1982. In her entire Wimbledon career of 22 competitions, King never failed to be a semi-finalist in at least one event every year. Evert repeated her Wimbledon quarterfinal victory over King at the clay court US Open, winning 6–2, 6–0. This loss prompted King to say, "I better get it together by October or November or that's it. I'll have to make some big decisions. I'm not 20-years-old and I can't just go out and change my game. It's only the last four weeks I haven't been in [knee] pain. [But if] I keep using that as a copout, I shouldn't play." The remainder of the year, King's win–loss record was 31–3, losing only to Evert, Dianne Fromholtz Balestrat, and Michelle Tyler Wilson. King won five of the eight tournaments she entered plus both of her Wightman Cup matches. She defeated Navratilova all four times they played, including three times in three consecutive weeks, and beat Wimbledon champion Virginia Wade twice. Beginning September 26, King played seven consecutive weeks. She lost to Tyler in the second </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> round in Palm Harbor, Florida and Fromholtz Balestrat in the semifinals in Atlanta. She then won three hard court tournaments in three consecutive weeks. She defeated Navratilova and Wendy Turnbull to win in Phoenix, losing only four points to Turnbull in the third set of the final. The next week, she defeated Navratilova, Fromholtz Balestrat, and Wimbledon runner-up Stöve to win in São Paulo. The third week, she defeated Ruzici, Stöve, and Janet Newberry Wright to win in San Juan. In November, Evert snapped King's 18-match winning streak in the final of the Colgate Series Championships in Mission Hills, California. King then won her Wightman Cup matches, defeated Navratilova to win the tournament in Japan, and beat Wade to win the Bremar Cup in London. King said, "I have never had a run like this, even in the years when I was Wimbledon champion. At 34, I feel fitter than when I was 24." 1978 King played ten singles tournaments during the first half of 1978, limiting herself to doubles after Wimbledon. To start the year, King was the runner-up in Houston and Kansas City (losing to Martina Navratilova in both) and in Philadelphia (losing to Chris Evert). At the Virginia Slims Championships, King lost her first round robin match to Virginia Wade and defaulted her two remaining round robin matches because of a leg injury sustained during the first match. At Wimbledon, King played with a painful heel spur in her left foot and lost to Evert in the quarterfinals for the second consecutive year 6–3, 3–6, 6–2. The match was on-serve in the third set with King serving at 2–3 (40–0) before Evert won five consecutive points to break serve. King won a total of only two points during the last two games. King said after the match, "I don't think my mobility is very good and that's what I need to beat her. Physically, she [Evert] tears your guts apart unless you can stay with her. I'm really disappointed. I really wanted to play well. I just couldn't cut it because of my heel." King and her partner Ray Ruffels lost in the final of mixed doubles in straight sets. King teamed with Navratilova to win the women's doubles title at the US Open, King's fourth women's doubles title at that tournament and fourteenth Grand Slam women's doubles title overall. To end the year, King was undefeated in five doubles matches (four with Evert and one with Rosemary Casals) as the U.S. won the Federation Cup in Melbourne, Australia. During the Federation Cup competition, King hinted at retirement from future major singles competitions and said that she was "sick and tired of continued surgery" in trying to get fit enough for those events. Nevertheless, King had foot surgery on December 22 in an attempt to regain mobility for a return to the tennis tour. 1979 During the first half of 1979, King played only one event – doubles in the Federation Cup tie against Spain – because of major surgery to her left foot during December 1978. King returned to singles competition at the Wimbledon warm-up tournament in Chichester. She defeated the reigning Wimbledon champion, Martina Navratilova, in a 48-minute quarterfinal 6–1, 6–2 before losing to Evonne Goolagong Cawley in the semifinals 1–6, 6–4, 10–8. Seeded seventh at Wimbledon, King defeated Hana Mandlíková in the fourth round before losing the last six games of the quarterfinal match with fourth-seeded Tracy Austin 6–4, 6–7(5), 6–2. King partnered with Navratilova at Wimbledon to win King's 20th and final Wimbledon title, breaking Elizabeth Ryan's longstanding record of 19 Wimbledon titles </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> just one day after Ryan collapsed and died at Wimbledon. At the US Open, the ninth-seeded King reached the quarterfinals without dropping a set, where she upset the fourth-seeded Virginia Wade 6–3, 7–6(4). Next up was a semifinal match with the four-time defending champion and top-seeded Chris Evert; however, with King hampered by a neck injury sustained during a bear hug with a friend the day before the match, Evert won 6–1, 6–0, including the last eleven games and 48 of the last 63 points. This was Evert's eighth consecutive win over King, with Evert during those matches losing only one set and 31 games and winning four 6–0 sets. Evert said after the match, "Psychologically, I feel very confident when I ... play her." The following week in Tokyo, King won her first singles title in almost two years, defeating Goolagong Cawley in the final. In November in Stockholm, King defeated Betty Stöve in the final after Stöve lost her concentration while serving for the match at 5–4 in the third set. Three weeks later in Brighton, King lost a semifinal match with Navratilova 7–5, 0–6, 7–6(3) after King led 6–5 in the third set. She ended the year with a quarterfinal loss in Melbourne (not the Australian Open), a second round loss in Sydney, and a three-set semifinal loss to Austin in Tokyo. 1980–1981 King won the tournament in Houston that began in February, snapping Martina Navratilova's 28-match winning streak in the straight-sets final. At the winter series-ending Avon Championships in March, King defeated Virginia Wade in her first round robin match 6–1, 6–3. After Wade held serve at love to open the match, King won nine consecutive games and lost only nine points during those games. King then lost her second round robin match to Navratilova and defeated Wendy Turnbull in an elimination round match, before losing to Tracy Austin in the semifinals King played the 1980 French Open, her first time since she won the event in 1972 and completed a career singles Grand Slam. She was seeded second but lost in the quarterfinals to fifth-seeded Dianne Fromholtz Balestrat of Australia. At Wimbledon, King defeated Pam Shriver in a two-hour, forty minute fourth round match after King saved a match point in the second set and recovered from a 4–2 (40–0) deficit in the third set with Shriver serving. In a quarterfinal that took two days to complete, King lost to two-time defending champion and top-seeded Navratilova 7–6, 1–6, 10–8. The beginning of the match was delayed until late afternoon because of rain. Because she wore glasses, King agreed to start the match then on condition that tournament officials immediately suspend the match if the rain resumed. During the first set, drizzle began to fall; however, the chair umpire refused to suspend the match. King led in the tiebreaker 5–1 before Navratilova came back to win the set, whereupon the umpire then agreed to the suspension. When the match resumed the next day, King won 20 of the first 23 points to take a 5–0 lead in the second set and lost a total of seven points while winning the set in just 17 minutes. In the third set, Navratilova broke serve to take a 2–0 lead before King broke back twice and eventually served for the match at 6–5. King then hit four volley errors, enabling Navratilova to break serve at love and even the match. King saved three match points while serving at 6–7 and three more match points while serving at 7–8. During the change-over between games at 8–9, King's glasses broke for </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> the first time in her career. She had a spare pair, but they did not feel the same. King saved two match points before Navratilova broke serve to win the match. King said, "I think that may be the single match in my career that I could have won if I hadn't had bad eyes." King teamed with Navratilova to win King's 39th and final Grand Slam title at the US Open. Navratilova then decided she wanted a new doubles partner and started playing with Shriver but refused to discuss the change directly with King. She finally confronted Navratilova during the spring of 1981, reportedly saying to her, "Tell me I'm too old ... but tell me something." Navratilova refused to talk about it. King had minor knee surgery on November 14 in San Francisco to remove adhesions and cartilage. 1982–1983 In 1982, King was 38 years old and the twelfth-seed at Wimbledon. In her third round match with Tanya Harford of South Africa, King was down 7–5, 5–4 (40–0) before saving three match points to win the second set 7–6(2) and then the third set 6–3. King said in her post-match press conference, "I can't recall the previous time I have been so close to defeat and won. When I was down 4–5 and love–40, I told myself, 'You have been here 21 years, so use that experience and hang on. In the fourth round, King upset sixth-seeded Australian Wendy Turnbull in straight sets. King then upset third-seeded Tracy Austin in the quarterfinals 3–6, 6–4, 6–2 to become the oldest female semifinalist at Wimbledon since Dorothea Douglass Lambert Chambers in 1920. This was King's first career victory over Austin after five defeats and reversed the result of their 1979 Wimbledon quarterfinal. King said in her post-match press conference, "Today, I looked at the scoreboard when I was 2–0 in the third set and the '2' seemed to be getting bigger and bigger. In 1979, when I was up 2–0 at the same stage, I was tired and didn't have anything left. But today I felt so much better and was great mentally." Two days later in the semifinals, which was King's 250th career match at Wimbledon in singles, women's doubles, and mixed doubles, the second-seeded Chris Evert defeated King on her fifth match point 7–6(4), 2–6, 6–3. King was down a set and 2–1 in the second set before winning five consecutive games to even the match. King explained that she actually lost the match in the first set by failing to convert break points at 15–40 in the second and fourth games. She reached the semifinals in her final appearance at Wimbledon, losing to Andrea Jaeger 6–1, 6–1 after beating Kathy Jordan in the quarterfinals, seventh-seeded Wendy Turnbull in the fourth round, and Rosemary Casals, her longtime doubles partner, in the third round. Jaeger claims that she was highly motivated to defeat King because King had defeated Turnbull, a favorite of Jaeger's, and because King refused a towel from an attendant just before her match with Jaeger, explaining, "I'm not going to sweat in this match." King became the oldest WTA player to win a singles tournament when she won the Edgbaston Cup grass court tournament in Birmingham at 39 years, 7 months and 23 days after a straight-sets victory in the final against Alycia Moulton. The final official singles match of King's career was a second round loss to Catherine Tanvier at the 1983 Australian Open. 1984 to present King played doubles sporadically from 1984 through 1990. She retired from competitive play in doubles in March 1990. </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> In her last competitive doubles match, King and her partner, Jennifer Capriati, lost a second round match to Brenda Schultz-McCarthy and Andrea Temesvári 6–3, 6–2 at the Virginia Slims of Florida tournament. King became the captain of the United States Fed Cup team and coach of its women's Olympic tennis squad. She guided the U.S. to the Fed Cup championship in 1996 and helped Lindsay Davenport, Gigi Fernández, and Mary Joe Fernández capture Olympic gold medals. In 2002, King dismissed Capriati from the Fed Cup team, saying Capriati had violated rules that forbade bringing along and practicing with personal coaches. Opinion was sharply divided, with many supporting King's decision but many feeling the punishment was too harsh, especially in hindsight when Monica Seles and Lisa Raymond were defeated by lower-ranked Austrians Barbara Schett and Barbara Schwartz. The following year, Zina Garrison succeeded King as Fed Cup captain. Activism within the tennis profession Player compensation Before the start of the open era in 1968, King earned US$100 a week as a playground instructor and student at California State University, Los Angeles when not playing in major tennis tournaments. In 1967, King criticized the United States Lawn Tennis Association (USLTA) in a series of press conferences, denouncing what she called the USLTA's practice of "shamateurism", where top players were paid under the table to guarantee their entry into tournaments. King argued that this was corrupt and kept the game highly elitist. King quickly became a significant force in the opening of tennis to professionalism. King said this about the amateur game, Push for gender equality When the open era began, King campaigned for equal prize money in the men's and women's games. In 1971, her husband, Larry King created the idea to form a nine player women's group with the financial backing of World Tennis magazine founder, Gladys Heldman, and the sponsorship of Virginia Slims Chairman Joe Cullman. King became the first woman athlete to earn over US$100,000 in prize money; however, inequalities continued. King won the US Open in 1972 but received US$15,000 less than the men's champion Ilie Năstase. She stated that she would not play the next year if the prize money were not equal. In 1973, the US Open became the first major tournament to offer equal prize money for men and women. King led player efforts to support the first professional women's tennis tour in the 1970s called the Virginia Slims, founded by Gladys Heldman and funded by Joseph Cullman of Philip Morris. Once the tour took flight, King worked tirelessly to promote it even though many of the other top players were not supportive. "For three years we had two tours and because of their governments [Martina] Navratilova and Olga Morozova had to play the other tour. Chris [Evert], Margaret [Court], Virginia [Wade], they let us do the pioneering work and they weren't very nice to us. If you go back and look at the old quotes; they played for the love of the game, we played for the money. When we got backing and money, we were all playing together – I wonder why? I tried not to get upset with them. Forgiveness is important. Our job was to have one voice and win them over." In 1973, King became the first President of the women's players union – the Women's Tennis Association. In 1974, she, with husband Larry King and Jim Jorgensen, founded womenSports magazine and started the Women's Sports Foundation. Also in 1974, World TeamTennis began, founded by Larry King, Dennis Murphy, Frank Barman and Jordan Kaiser. She became league commissioner in 1982 </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> and major owner in 1984. King is a member of the Board of Honorary Trustees for the Sports Museum of America, which opened in 2008. The museum is the home of the Billie Jean King International Women's Sports Center, a comprehensive women's sports hall of fame and exhibit. Other activities King's husband Larry co-founded World Team Tennis in 1973 with Dennis Murphy, Jordan Kaiser, and Fred Barman and WTT began in 1974. The couple used their savings to put on a team tennis event at the Oakland Coliseum. King remained involved with World Team Tennis for decades, eventually sharing ownership with her ex-husband, her life partner Ilana Kloss and USTA. In 2017, King sold her majority ownership stake of the league to Mark Ein and Fred Luddy. WTT was based on her philosophy for gender equality and it had been running continuously for over 40 years. In 1999, King was elected to serve on the Board of Directors of Philip Morris Incorporated, garnering some criticism from anti-tobacco groups. She no longer serves in that capacity. King was involved in the Women's Sports Foundation and the Elton John AIDS Foundation. She also served on the President's Council for Fitness, Sports and Nutrition as a way to encourage young people to stay active In 2008, King published the book Pressure is a Privilege: Lessons I've Learned from Life and the Battle of the Sexes. In December 2013, US President Barack Obama appointed King and openly gay ice hockey player Caitlin Cahow to represent the United States at the 2014 Winter Olympics in Sochi, Russia. This has been interpreted as a signal on gay rights, in the context of concerns and controversies at the 2014 Winter Olympics regarding LGBT rights in Russia. King was forced to drop out of the delegation due to her mother's ill health. Betty Moffitt, King's mother, died on February 7, 2014, the day of the 2014 Winter Olympics opening ceremonies. Billie Jean was selected to deliver the Northwestern University commencement address on June 16, 2017 in Evanston, Illinois. She attended the 75th Golden Globe Awards in 2018 as a guest of Emma Stone. King and Kloss became minority owners of the Los Angeles Dodgers baseball team in September 2018. Awards, honors, and tributes Tributes from other players Margaret Court, who won more Grand Slam titles than anyone, has said that King was "the greatest competitor I've ever known". Chris Evert, winner of 18 Grand Slam singles titles, has said, "She's the wisest human being that I've ever met and has vision people can only dream about. Billie Jean King is my mentor and has given me advice about my tennis and my boyfriends. On dealing with my parents and even how to raise children. And she doesn't have any." In 1979, several top players were asked who they would pick to help them recover from a hypothetical deficit of 1–5 (15–40) in the third set of a match on Wimbledon's Centre Court. Martina Navratilova, Rosemary Casals, and Françoise Dürr all picked King. Navratilova said, "I would have to pick Billie Jean at her best. Consistently, Chris [Evert] is hardest to beat but for one big occasion, one big match, one crucial point, yes, it would have to be Billie Jean." Casals said, "No matter how far down you got her, you never could be sure of beating her." Awards and honors King was the Associated Press Female Athlete of the Year in 1967. In 1972, King became the first tennis player to be named Sports Illustrated Sportsman of the Year. She was also the first female athlete ever </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> to receive that honor. In 1975, Seventeen magazine found that King was the most admired woman in the world from a poll of its readers. Golda Meir, who had been Israel's prime minister until the previous year, finished second. In a May 19, 1975, Sports Illustrated article about King, Frank Deford noted that she had become something of a sex symbol. King was inducted into the International Tennis Hall of Fame in 1987. Life magazine in 1990 named her one of the "100 Most Important Americans of the 20th Century". King was the recipient of the 1999 Arthur Ashe Courage Award. In 1999 King was inducted into the Chicago Gay and Lesbian Hall of Fame. In 2000, King received an award from GLAAD, an organization devoted to reducing discrimination against gays, lesbians, bisexuals, and transgender people, for "furthering the visibility and inclusion of the community in her work". In 2006, the Women's Sports Foundation began to sponsor the Billie Awards, which are named after and hosted by King. On August 28, 2006, the USTA National Tennis Center in Flushing Meadows-Corona Park was rededicated as the USTA Billie Jean King National Tennis Center. John McEnroe, Venus Williams, Jimmy Connors, and Chris Evert were among the speakers during the rededication ceremony. In 2006, California Governor Arnold Schwarzenegger and his wife Maria Shriver inducted King into the California Hall of Fame located at The California Museum for History, Women, and the Arts. On November 20, 2007, King was presented with the 2007 Sunday Times Sports Women of the Year Lifetime Achievement award for her contribution to sport both on and off the court. She was honored by the Office of the Manhattan Borough President in March 2008 and was included in a map of historical sites related or dedicated to important women. On August 12, 2009, President Barack Obama awarded King the Presidential Medal of Freedom for her work advocating for the rights of women and the lesbian, gay, bisexual, and transgender community. She was inducted into the Southern California Tennis Hall of Fame on August 5, 2011. On August 2, 2013, King was among the first class of inductees into the National Gay and Lesbian Sports Hall of Fame. In 2014, she was named one of ESPNW's Impact 25. King was shown in Marie Claire magazine's "The 8 Greatest Moments for Women in Sports". King received the BBC Sports Personality of the Year Lifetime Achievement Award on December 16, 2018. It was presented to by long-time friend and fellow tennis player and broadcaster Sue Barker, making King only the second American (after Michael Phelps) and the first American woman to win the award. Cal State LA's more than athletic facility is named the Billie Jean King Sports Complex. The sports complex—which was approved by the California State University Board of Trustees on September 21—features the Eagle's Nest Arena, the University Stadium, Jesse Owens Track and Field, Reeder Field (baseball), the swimming pool, and tennis and basketball courts. Playing style and personality King learned to play tennis on the public courts of Long Beach, California, and was coached by tennis teacher Clyde Walker. She furthered her tennis career at the Los Angeles Tennis Club. She was an aggressive, hard-hitting net-rusher with excellent speed, Chris Evert, however, said about King, "Her weakness is her impatience." Concerning her motivations in life and tennis, King said, In a 1984 interview, just after she had turned 40, King said, Julie Heldman, who frequently played King but never felt close to her, said about King's personality, Kristien Shaw, another frequent opponent of King, said, Concerning the qualities of </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> a champion tennis player, King said, In popular culture King's friend Elton John wrote the song "Philadelphia Freedom", a nod to her World TeamTennis team, for King. The song was released New Year's Day 1975 and became a number one hit. Charles M. Schulz, creator of the Peanuts comic strip, was an admirer and close friend. Schulz referred to King several times in Peanuts and used the comic strip to support the women's sports movement after becoming friends with King. Actress Holly Hunter portrayed King in the 2001 ABC television film When Billie Beat Bobby. King played a judge on Law & Order in 2007, and appeared as herself on The Odd Couple in 1973, Ugly Betty in May 2009, and Fresh Off The Boat in 2016. Actress Emma Stone portrayed King in the 2017 biopic Battle of the Sexes. The Ted Tinling-designed dress King wore for the real Battle of the Sexes match is part of a Smithsonian Museum collection. Personal life Billie Jean and Larry King were engaged in fall of 1964 and married in Long Beach, California, on September 17, 1965. Billie Jean credited Larry with introducing her to feminism and for pushing her to pursue tennis as a career. Billie Jean later said she "was totally in love with Larry" when they married. By 1968, King realized that she was attracted to women, and in 1971, began an intimate relationship with her secretary, Marilyn Barnett (born Marilyn Kathryn McRae on January 28, 1948). Billie Jean acknowledged the relationship when it became public in a May 1981 palimony lawsuit filed by Barnett, making Billie Jean the first prominent professional female to come out. Feeling she could not admit to the extent of the relationship, Billie Jean publicly called it a fling and a mistake. She remained married to Larry. The lawsuit caused Billie Jean to lose an estimated $2 million in endorsements and forced her to prolong her tennis career to pay attorneys. Also in 1971, she had an abortion that was made public in a Ms. magazine article. Larry had revealed Billie Jean's abortion without consulting her. Concerning the personal cost of concealing her sexuality for so many years, Billie Jean said: Billie Jean and Larry remained married through the palimony suit fallout. The marriage ended in 1987 after Billie Jean fell in love with her doubles partner, Ilana Kloss. Billie Jean and Larry remained on good terms, with Billie Jean serving as godmother to Larry's son from his subsequent marriage. Billie Jean King has residences in New York City and Chicago with Kloss, her life partner. Grand Slam statistics Grand Slam single finals 18 finals (12 titles, 6 runner-ups) Grand Slam tournament timeline Singles Note: The Australian Open was held twice in 1977, in January and December. See also List of Grand Slam women's singles champions List of Wimbledon mixed doubles champions List of French Open mixed doubles champions List of US Open mixed doubles champions List of French Open women's singles champions List of Wimbledon ladies' singles champions List of US Open women's singles champions List of French Open women's doubles champions List of Wimbledon ladies' doubles champions List of US Open women's doubles champions List of Open Era tennis records All-time tennis records – women's singles WTA Tour records Open Era tennis records – women's singles Homosexuality in sports in the United States Billie Jean King Leadership Initiative Notes References Further reading Jones, Ann, A Game of Love, 1971 Overman, Steven J. and K. B. Sagert, Icons of Women's Sport. Greenwood Press, 2012, Vol. 1. Ware, Susan (2011). Game, Set, Match: Billie Jean </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> King and the Revolution in Women's Sports. University of North Carolina Press. Combines biography and history in a study of the tennis player, liberal feminism, and Title IX. External links Category:Billie Jean King Category:1943 births Category:American female tennis players Category:American feminists Category:American Methodists Category:Australian Championships (tennis) champions Category:Los Angeles State College alumni Category:French Championships (tennis) champions Category:French Open champions Category:Grand Slam (tennis) champions in mixed doubles Category:Grand Slam (tennis) champions in women's doubles Category:Grand Slam (tennis) champions in women's singles Category:International Tennis Hall of Fame inductees Category:Lesbian feminists Category:Lesbian sportswomen Category:LGBT people from California Category:LGBT sportspeople from the United States Category:LGBT tennis players Category:Living people Category:Presidential Medal of Freedom recipients Category:Sportspeople from Long Beach, California Category:Tennis commentators Category:Tennis people from California Category:Toastmasters members Category:United States National champions (tennis) Category:US Open (tennis) champions Category:Wimbledon champions Category:Women sports announcers Category:World No. 1 tennis players Category:Women's Sports Foundation Category:BBC Sports Personality Lifetime Achievement Award recipients Category:Los Angeles Dodgers executives Category:Los Angeles Dodgers owners Category:Major League Baseball executives </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Cyaniris Cyaniris is a genus of butterflies in the family Lycaenidae. Recent molecular studies have determined that Cyaniris is a different genus from Polyommatus, where it has been included for several years. Species Two species are recognized: Cyaniris semiargus (Rottemburg, 1775) Cyaniris bellis (Freyer, 1845) References Category:Polyommatini Category:Lycaenidae genera </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Cedarville School Cedarville School is a historic school building located at Cedarville, Gilmer County, West Virginia. It was built in 1923, and is a two-story "T"-shaped, wood frame building with a hipped roof, measuring approximately 50 feet deep and 80 feet wide. It has striated stuccoed exterior surfaces and is on a raised basement of locally quarried stone. It was used as a school until 1968, after which it was converted to apartments. It was listed on the National Register of Historic Places in 1994. References Category:American Craftsman architecture in West Virginia Category:American Foursquare architecture in West Virginia Category:Apartment buildings in West Virginia Category:Bungalow architecture in West Virginia Category:Defunct schools in West Virginia Category:Educational institutions disestablished in 1968 Category:Educational institutions established in 1923 Category:Former school buildings in the United States Category:National Register of Historic Places in Gilmer County, West Virginia Category:School buildings completed in 1923 Category:1923 establishments in West Virginia Category:School buildings on the National Register of Historic Places in West Virginia Category:1968 disestablishments in West Virginia </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Kambatta Viswanathar Temple, Kumbakonam Kambatta Viswanathar Temple is a Hindu temple dedicated to Shiva located at Kumbakonam in Thanjavur district, Tamil Nadu, India. Presiding deity The moolavar presiding deity, is found in his manifestation as Visveswar. His consort, Parvati, is known as Anandhanithi. Specialty 12 Shiva temples are connected with Mahamaham festival which happens once in 12 years in Kumbakonam. They are : Kasi Viswanathar Temple, Kumbeswarar Temple, Someswarar Temple, Nageswara Temple, Kalahasteeswarar Temple, Gowthameswarar Temple, Kottaiyur Kodeeswarar temple Amirthakalasanathar Temple, Banapuriswarar Temple, Abimukeswarar Temple, Kumbakonam, Kambatta Visvanathar Temple and Ekambareswarar Temple. This temple is one among them. Mint During the period of Cholas, there was mint in this place. So this place was called as 'Kambattam' (in Tamil), the place where gold and silver coins were produced. Mahasamprokshanam The Mahasamprokshanam also known as Kumbabishegam of the temple was held on 26 October 2015. See also Hindu temples of Kumbakonam Mahamaham References External links Mahasamprokshanam 26 October 2015 Category:Hindu temples in Kumbakonam Category:Shiva temples in Thanjavur district </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Mandarin Chinese in the Philippines Mandarin Chinese is spoken and taught academically to Chinese Filipinos in the Philippines. Both Standard Chinese and Taiwanese Mandarin are taught and spoken in the Philippines, with some schools and speakers using Simplified Chinese characters, some using Traditional Chinese characters, and some using a mixture of both. Classification Mandarin in the Philippines can be classified into two distinct Mandarin dialects: Standard Mandarin and Colloquial Mandarin. Standard Mandarin is either the standard languages of Mainland China and Taiwan, while Colloquial Mandarin in the Philippines tends to combine features from Mandarin () and features from Hokkien () of the local Philippine Hokkien dialect, which is the heritage language of many Chinese Filipinos. Usage Only a small minority of Chinese Filipinos claim Mandarin as their native first language, with Tagalog or English typically being the first language. The lack of environment for speaking the language and the difficulty of learning it created not just a lack of interest, but even great disgust by some towards it. Efforts in the 21st century to promote Mandarin Chinese education in Chinese Filipino institutions and recent utilitarian trends, such as more Mandarin job opportunities, recent immigrants from China or Taiwan, summer education trips to China or Taiwan, encouragement of universities and schools by past presidents, and education exchange deals with China have spurred interest and potential for growth in the usage of Mandarin. Code-switching Sometimes Chinese Filipinos also code-switch Mandarin together with other languages, such as English, Tagalog (or other Philippine languages), and Hokkien, as a form of pidgin language, just like Hokaglish or Singlish. Education In terms of phonology, vocabulary and grammar, the Mandarin taught in the Philippines is often the Taiwanese variety ("Guóyǔ") of Standard Chinese because many Chinese Filipino schools use dictionaries and books from Taiwan. In recent years, some have also began using books and teaching materials from Mainland China, Singapore, and Malaysia. Filipino Mandarin newspapers use the Traditional Chinese characters in writing. Due to selection of founders and sponsors of Chinese schools, schools either teach using only Simplified Chinese characters, only Traditional Chinese characters, or a mixture of both. 'Many Chinese Filipino schools use pinyin or bopomofo (zhuyin fuhao) to teach the language. Chinese Filipino schools often use the first language approach, which assumes that students of Chinese Filipino schools have had native experience of Mandarin. See also Philippine Hokkien Languages of Philippines References Category:Chinese-Filipino culture Category:Mandarin Chinese * </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Alfred Ollivant (writer) Alfred Ollivant (1874–1927) was an English novelist known best for his novel Owd Bob. Ollivant also published about a dozen other novels ranging from small-scale cautionary tales to grand historical stories. Alfred Ollivant was born in Nuthurst, Sussex, in 1874 and became an author after a horse-riding injury ended his brief military career. Owd Bob, his first novel, was published in 1898. Set in rural Cumbria, in northern England, the novel concerns a suspected sheepkilling collie, Bob. Even though most of the book's dialogue is written in the Cumbrian dialect, it became popular in the United States. Ollivant published a sequel, Danny, in 1902. He was also a short story contributor to the magazines The Atlantic Monthly and the Boston Evening Transcript. Alfred Ollivant died in London on 19 January 1927. Bibliography Owd Bob - Being the story of Bob, son of Battle, the last of the grey dogs of Kenmuir (1898) Danny (1902) The Redcoat Captain: A Story of that Country (1907) The Gentleman: A Romance of the Sea (1908) The Taming of John Blunt (1911) The Royal Road (1912) The Brown Mare, and other studies of England under the cloud (1916) Boy Woodburn: A story of the Sussex Downs (1917) Two Men: A romance of Sussex (1919) One Woman: Being the second part of a romance of Sussex (1921) Devil Dare (1923) "Old For-ever": an epic of beyond the Indus (1923) Boxer & Beauty: A tale of two cart-horses (1924) Tomorrow (1927) The Next Step, an essay on the missing policeman dedicated to the masses of mankind, silent and suffering. (1919) External links Category:20th-century English novelists Category:1927 deaths Category:1874 births Category:English male novelists Category:20th-century British male writers </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Illinois Reserve Militia The Illinois Reserve Militia was the state defense force of Illinois during World War I and World War II. After the Illinois National Guard was federalized, the Illinois Reserve Militia was organized to assume the stateside duties of the National Guard. History World War I Before the Illinois National Guard was called into federal service, the State Council of Defense established the Illinois Volunteer Training Corps in order to begin training citizen volunteers in paramilitary skills. On 25 June 1917, legislation creating the Illinois Reserve Militia was signed. By 9 November 1918 there were 8 regiments of Reserve Militia, plus 1 separate battalion, and 38 separate companies, organized in two brigades, with approximately 475 officers and 7,000 enlisted men. In August 1919, approximately 1,500 soldiers from the 7th and 10th regiments of the Illinois Reserve Militia were stationed at their headquarters in Peoria, Illinois, to be prepared to respond to potential rioting from striking workers at the Keystone Steel and Wire Company. World War II The Illinois Reserve Militia was reactivated in 1941, and by December 1941, the Reserve Militia was on continuous duty. Illinois organized the Reserve Militia as a full infantry division and an air corps, totaling nearly 6,000 soldiers by June 1944, with its members agreeing to serve a two-year initial enlistment. In September 1942, the Illinois Reserve Militia was mobilized and placed on standby during a strike by employees of the Western Cartridge Company. In December 1945, approximately 450 members of the Illinois Reserve Militia were mobilized to operate a shuttle service between all the major Chicago bus, train, and air terminals after various mass transit systems became overwhelmed due to the number of servicemen attempting to return to their homes. Legal status The authority of each state to maintain its own state defense force is recognized by the federal government of the United States under Title 32, Section 109 of the United States Code. Approximately twenty-three states and the territory of Puerto Rico currently maintain state defense forces. Under Illinois law, the Governor of Illinois has the legal authority to reactivate the Illinois Reserve Militia. See also Illinois Naval Militia Illinois Wing Civil Air Patrol References Category:Military in Illinois Category:State defense forces of the United States </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Ian Boyton Ian Justin Boyton (born 19 August 1974) is a former English cricketer. Boyton was a right-handed batsman who bowled right-arm medium pace. He was born in Barking, London. Boyton represented the Middlesex Cricket Board in a single List A match against Scotland in the 1st round of the 2002 Cheltenham & Gloucester Trophy which was held in 2001. In his only List A match he scored 31 runs and with the ball he took a single wicket at a cost of 21 runs. References External links Ian Boyton at Cricinfo Ian Boyton at CricketArchive Category:1974 births Category:Living people Category:People from Barking, London Category:Cricketers from Greater London Category:English cricketers Category:Middlesex Cricket Board cricketers </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Folwarki, Świętokrzyskie Voivodeship Folwarki is a village in the administrative district of Gmina Stopnica, within Busko County, Świętokrzyskie Voivodeship, in south-central Poland. It lies approximately north of Stopnica, east of Busko-Zdrój, and south-east of the regional capital Kielce. References Category:Villages in Busko County </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.
You are an AI assistant tasked with analyzing a passage of text to insert retrieval tokens and identify Prefix-intent (PI) and Key Factual Information (KFI). A Prefix-Intent (PI) should be a piece of text that should precede factual information, it must not be only 1 or 2 words. Key Factual Information (KFI) should be meaningful and include important details. Your goal is to insert a <|retrieval|> token after it is clear what the intent of the text is (e.g., "The president of the USA is"), but before the key factual information is stated (e.g., "Joe Biden"). Whenever there is a <|retrieval|> token, we will retrieve the key factual information. Here is the passage you will be working with: <passage> Return of the Wild Horses Return of the Wild Horses is the most important in-situ conservation project organized by Prague Zoo. Its aim is to increase the numbers and genetic diversity of Przewalski's horses in their native habitat in Mongolia. Resuming previous similar European projects, Return of the Wild Horses transports and reintroduces captive-bred wild horses into large nature reserves in Mongolia, the project also provides material support to Mongolian rangers who protect and monitor the herds in Great Gobi B Strictly Protected Area. Through yearly transports of horses from the Czech Republic starting in 2011, Zoo Prague managed to increase the population of wild horses in Khomyn Tal by 34 animals. Development of the project Prague Zoo manages the international Przewalski's horse studbook since 1959, in 1997 it joined the reintroduction efforts in Mongolia, contributing four of its horses to Takhin Tal Nature Reserve and Khustain Nuruu National Park. Prague Zoo exhibits and breeds horses in Prague and keeps a core herd in its Przewalski's Horse Breeding and Acclimatization Facility in Dolní Dobřejov (Střezimíř municipality). The management office of Great Gobi B Strictly Protected Area was opened in 2006 and began cooperating with Prague Zoo, who continues to provide important material and technical support to park management and guards such as satellite phones, GPS navigation, microscopes, binoculars, off-road cars, motorcycles, guardhouses, hay barns, diesel generator and a well. In 2007, Prague Zoo joined the International Takhi Group. The horses suitable for release are selected and transported from European zoos as well as from the core herd in Dobřejov. Once a year four of these are loaded in boxes, taken by car to airport, lifted by Czech Army cargo plane to Mongolia, then travel to their destination on lorries. After some time spent in acclimatization pen, they are released into the wild. The project is supported by Czech Army, Czech Development Agency, Czech and Mongolian embassies, public donations and proceeds from Prague Zoo ticket sales. Yearly transports of horses The first four Przewalski's horses were transported to Khomyn Tal in 2011. All following transports have been headed to Great Gobi B Strictly Protected Area. The total number of reintroduced horses from Czech Republic so far is 34. Four more horses were transported by Prague Zoo within Mongolia during the eighth Return of the Wild Horses, from Khustain Nuruu National Park near Ulaanbaatar to Takhin Tal Nature Reserve in Gobi B. Great Gobi B Strictly Protected Area is now home to more than 270 Przewalski's horses, the transported horses have formed or joined the local herds and reproduce successfully. See also Wildlife conservation Reintroduction in-situ conservation References External links Documentary Návrat divokých koní 2011 at the website of Česká televize, in Czech Documentary Návrat divokých koní 2012 at youtube.com Return of the Wild Horses website of Tierpark Berlin Category:Zoos in the Czech Republic Category:Conservation projects Category:Wildlife conservation Category:Animal reintroduction Category:Fauna of East Asia </passage> Follow these steps to complete the task: 1. Inserting retrieval tokens: a. Read through the text carefully. b. Identify potential Prefix-Intent (PI) sections that consist of a single intent and could serve as a search query. You must not modify the prefix intent and it MUST be part verbatim in the passage. It cannot be single or only a few words. c. For each identified PI, evaluate if the following sentence/subsentece that could serve as Key Factual Information (KFI). d. Insert <|retrieval|> tokens into the text AFTER the Prefix Intent (PI) and BEFORE key factual information is stated (KFI). e. Key Factual information should not intersect with Prefix Intent. 2. Identifying Prefix-Intent (PI) and Key Factual Information (KFI): a. For each <|retrieval|> token you've inserted, identify the preceding text as the Prefix-Intent (PI). b. Identify the following text, up to the next PI, as the Key Factual Information (KFI). c. Ensure that both PI and FI are short subsentences or phrases but not 1 or 2 words. 3. Insert <|start-PI|>,<|end-PI|>, <|start-KFI|>, <|end-KFI|>. a. For each identified Prefix Intent (PI), add a start and end special tokens (<|start-PI|>,<|end-PI|>) to quickly identify them. b. Do the same for the Key Factual Information (KFI) (<|start-KFI|>, <|end-KFI|>) c. Ensure that the very first PI in the passage is preceded by <|start-PI|>. d. Double-check that every PI and KFI is properly enclosed in these tags. 4. Describing ideal retrieved information: a. Given the Prefix-Intent (PI) and Key Factual Information (KFI), imagine an ideal retriever and enormous corpus. b. Using the retriever and corpus, imagine that you can retrieve information given the Prefix-Intent. c. Describe how that content should look like so it helps generating the Key Factual Information (KFI) 5. Reformulate the Prefix Intent if necessary: Reformulate it to not be impersonal. This means that if the subject it is not identified (e.g. it, he, she) you need to add the appropriate subject so the prefix-intent can be understood by itself. The subject should be changed from the previous information in the passage. Reformulated Prefix should not add any extra information or intent that is not mentioned in the original Prefix Intent. 6. Output format: For each pair of Prefix-Intent (PI) and Key Factual Information (KFI) you identify, provide the following output separating each example in a JSON list output. Ensure that it is actually a JSON. - Prefix_intent: [Write the Prefix Intent here] - Prefix_reformulated: [Write the Prefix Intent reformulated] - Key_Factual_Information: [Write the Key Factual Information here] - Verification: [Confirm that the KFI can be directly derived from the PI given a strictly connected context] - Ideal_Knowledge: [Describe how the ideal retrieved content should look to generate the KFI] - Explanation: [Briefly explain how the PI and KFI connect and why you chose to insert the <|retrieval|> token in that position] 7. Final step: After you have completed the analysis and provided outputs for all identified PI-KFI pairs, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions Here's an example of the expected output format: Passage: Girmal Falls General This waterfall extends to a height of up to 100 feet, making it the highest waterfall of Gujarat. The picturesque beauty of this site makes it popular among visitors and people of the region alike. The water swiftly falls from a great height, creating a fog like condition that’s eye catching. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. The fall comes to its best form at the time of monsoon and provides an immensely striking appearance. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler. ```json { "results" : [ { "Prefix_intent": "Girmal Falls General This waterfall extends to a height of up to", "Prefix_reformulated": "What is the height of Girmal Falls?", "Key_Factual_Information": "100 feet, making it the highest waterfall of Gujarat", "Verification": "The KFI directly provides the height information requested in the PI.", "Ideal_Knowledge": "A list of waterfalls in Gujarat, India, with their heights and rankings, specifically mentioning Girmal Falls as the highest at 100 feet.", "Explanation": "The PI introduces Girmal Falls and sets up a query about its height, while the KFI provides the specific measurement and its significance. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the factual statement." }, { "Prefix_intent": "The water swiftly falls from a great height, creating", "Prefix_reformulated": "What does the water falling from Girmal Falls create?", "Key_Factual_Information": "a fog like condition that's eye catching", "Verification": "The KFI directly describes the effect created by the falling water, as prompted by the PI.", "Ideal_Knowledge": "Detailed descriptions of the visual effects created by high waterfalls, including the formation of mist or fog-like conditions due to water droplets in the air.", "Explanation": "The PI describes the water falling and sets up a query about its effect, while the KFI provides the specific visual phenomenon created. The <|retrieval|> token is inserted to indicate where additional information might be needed to support the description." }, { "Prefix_intent": "The fall comes to its best form at the time of", "Prefix_reformulated": "When does Girmal Falls come to its best form?", "Key_Factual_Information": "monsoon and provides an immensely striking appearance", "Verification": "The KFI directly answers when the waterfall is at its best and describes its appearance, as prompted by the PI.", "Ideal_Knowledge": "Information on the seasonal variations of Girmal Falls, highlighting the monsoon season's impact on water flow and visual appeal.", "Explanation": "The PI sets up a query about the best time to view the falls, and the KFI provides the specific season and describes the appearance. The <|retrieval|> token is inserted to indicate where additional information about seasonal changes might be needed." }], "Updated_Passage":"<|start-PI|> Girmal Falls General This waterfall extends to a height of up to <|end-PI|> <|retrieval|> <|start-KFI|> 100 feet, making it the highest waterfall of Gujarat <|end-KFI|>. The picturesque beauty of this site makes it popular among visitors and people of the region alike. <|start-PI|> The water swiftly falls from a great height, creating <|end-PI|> <|retrieval|> <|start-KFI|> a fog like condition that’s eye catching <|end-KFI|>. The government of this state is working on many projects to make this place an ideal picnic spot and a tourist attraction. <|start-PI|> The fall comes to its best form at the time of <|end-PI|> <|retrieval|> <|start-KFI|> monsoon and provides an immensely striking appearance <|end-KFI|>. Some of the best natural features of Gujarat can be explored in this place. This place is a nice and refreshing retreat for any traveler." } ``` 8. Final check: a. Review the Updated_Passage to ensure all PIs and KFIs are properly tagged. b. Verify that the first PI in the passage starts with <|start-PI|>. c. Make sure there are no missing or misplaced tags. Remember to analyze the entire passage, identify all relevant PI-KFI pairs, and provide the output for each pair as specified. After completing the analysis, present the original passage with the <|retrieval|> tokens inserted in their appropriate positions as part of a new key in the JSON file. Only Ouput the JSON File.