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>
Jayne Meadows Jayne Meadows (born Jane Cotter; September 27, 1919 – April 26, 2015), also known as Jayne Meadows-Allen, was an American stage, film and television actress, as well as an author and lecturer. She was nominated for three Emmy Awards during her career and was the elder sister of actress and memoirist Audrey Meadows. Early life Jayne Meadows was born Jane Cotter in 1919, in Wuchang, Wuhan, Hubei Province, China, the elder daughter of American Episcopal missionary parents, the Rev. Francis James Meadows Cotter and his wife, the former Ida Miller Taylor, who had married in 1915. Her younger sister was actress Audrey Meadows. She also had two brothers. In the early 1930s, the family settled in Sharon, Connecticut, where her father had been appointed rector of Christ Church. Career Meadows' most famous movies include: Undercurrent (with Katharine Hepburn), Song of the Thin Man (with William Powell and Myrna Loy), David and Bathsheba (with Gregory Peck, Susan Hayward and Raymond Massey), Lady in the Lake (with Robert Montgomery and Audrey Totter) and Enchantment (with David Niven and Teresa Wright). Louella Parsons presented to Meadows the Cosmopolitan Award for Finest Dramatic Performance of 1949, for Enchantment. Among her earliest television appearances, Meadows played reporter Helen Brady in a 1953 episode of Suspense opposite Walter Matthau titled, "F.O.B. Vienna". She was a regular panelist on the original version of I've Got a Secret and an occasional panelist on What's My Line?, the latter alongside husband Steve Allen. She also appeared on the NBC interview program Here's Hollywood. During the early days of the burgeoning live entertainment scene in Las Vegas, the Allens occasionally worked together as an act. Prior to Allen's death in 2000, the couple made several television appearances together. In 1998, they played an argumentative elderly couple in an episode of Homicide: Life on the Street in which Allen's character accidentally shoots a man in the act of committing suicide by plunging from the roof of the elderly couple's building. In 1999, the couple made their last joint TV appearance, (again playing a couple,) in the all-star episode of the Dick Van Dyke series Diagnosis: Murder, titled "The Roast", which marked Steve Allen's final screen appearance. She also appeared in City Slickers (as the telephone voice of Billy Crystal's character's over-protective and oversolicitous mother). Later life and death She was married to Steve Allen from 1954 until his death in 2000. They had one son, Bill. Allen's three children from his first marriage (Stephen Jr., Brian and David) are her stepchildren. Meadows was active in Republican affairs although Steve Allen was a Democrat. She was the recipient of several Doctor of Humane Letters degrees from various universities. Meadows remained active until 2009 when she fell and fractured her hip. Her last public appearance was in August 2009 at the Early TV Memories First-Class Commemorative Stamp Dedication Ceremony. She died on April 26, 2015, of natural causes at her home in Encino, California, aged 95. She is buried at Forest Lawn Memorial Park in the Hollywood Hills, beside Steve Allen. Filmography References External links Category:1919 births Category:2015 deaths Category:American Episcopalians Category:Actresses from Connecticut Category:American film actresses Category:American stage actresses Category:American television actresses Category:Pre-1949 Republic of China emigrants to the United States Category:American memoirists Category:Burials at Forest Lawn Memorial Park (Hollywood Hills) Category:People from Sharon, Connecticut Category:RCA Victor artists Category:California Republicans Category:Connecticut Republicans Category:Disease-related deaths in California Category:20th-century American actresses Category:21st-century American actresses Category:Women memoirists
</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>
Władysławów Bielawski Władysławów Bielawski is a village in the administrative district of Gmina Głowno, within Zgierz County, Łódź Voivodeship, in central Poland. It lies approximately north of Głowno, north-east of Zgierz, and north-east of the regional capital Łódź. References Category:Villages in Zgierz 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>
Chinese Underground Railroad The Chinese Underground Railroad was an imaginary route through the borderland between the United States and Mexico, especially around El Paso, Texas. Because of the 1882 Chinese Exclusion Act, Chinese immigrants, with the help of Chinese laborers living in Mexico and smugglers, would illegally enter the United States in order to bypass the act. Similar to the Underground Railroad that brought African-American slaves to free states in northern United States and Canada, the Chinese underground railroad was not actually a railroad. Additionally, the secret route allowed Chinese immigrants from the end of the 19th century to the beginning of the 20th century enter the United States during a time of discrimination against Chinese people. Background When the Chinese Exclusion Act was initially enacted in 1882, it allowed legal means for the Chinese to enter the United States. Those who came to travel, sell, or go to school were allowed to enter during the time of exclusion, and Chinese people already living in the United States were allowed to leave and come back so long as they had the proper certificate. However, when those who opposed Chinese immigration began to view the act as ineffective and ambiguous, violence against the Chinese rose. In order to placate the situation, Congress enacted the Scott Act of 1888, which no longer allowed Chinese workers to enter the United States with ease. After the Scott Act came the Geary Act in 1892, which extended the Chinese Exclusion Act for ten more years and mandated Chinese laborers receive certificates to prove their residence. While the Gresham-Yan Treaty of 1894 did allow Chinese laborers to leave and return to the United States if they had families, property, or debts in the United States, the exclusion law persisted until 1943. In order to get around the Chinese Exclusion Act, Chinese immigrants traveled to the United States through Canada and Latin America. The Railroad The underground railroad from Mexico to the United States was not actually a railroad, but it did follow the railroad lines between Sonora and Arizona. The underground railroad was found throughout the American-Mexican border, but its most common route was through El Paso. The route smuggled Chinese immigrants, as well as drugs (mainly opium) and diamonds to cities in the United States, such as Los Angeles to Chicago. Anglo-Americans and Chinese immigrants would team up to bring Chinese immigrants to the United States through Mexico. In the early 1900s, smugglers took advantage of the less populated regions in the borderland to bring Chinese immigrants into the United States. Smugglers would begin in Mexican cities such as Salina Cruz, Manzanillo, Mazatlán, and Guaymas and travel through Nogales, Arizona and Nogales Sonora, which were twin border cities, until they reached the United States. Traveling through the Mexican border could prove to be treacherous. Because it was largely a desert, many Chinese immigrants got lost easily and became weakened by dehydration and lack of food. Chinese immigrants who chose to rely on smugglers could find them untrustworthy - in one example, sixteen Chinese immigrants were killed after first collectively paying their smuggler guides. Today, the Chinese immigrants who suffered and died in the desert are remembered through the names of the desert. The area between San Felipe and the Mexicali-Imperial valleys is now called El Desierto de los Chinos (the Chinese desert), and other areas such as Chinaman Flat, El Chinero, and Llano El Chinero, commemorate the immigrant victims. The underground railroad was often supported by Chinese people who lived near the border. Chinese immigrants could move from Real de Castillo to Burro Valley to
</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>
Yalecites to Carisco thanks to the Chinese who lived in those villages. Those immigrants who successfully reached the American side of the border often continued to help other immigrants pass through the borderlands as well. One town that continued the underground railroad was El Paso Chinatown. One could illegally enter El Paso by going through Juárez, where Chinese immigrants learned English. Houses and buildings on the route concealed places for the Chinese immigrants to hide as they moved from one area to another, supported by other Chinese people in the area. However, the town was mostly filled with men who were unable to establish families because Chinese women were not allowed to enter the United States at all because of the Exclusion Act. Additionally, the Mexican Revolution of 1910 and the increased enforcing of the Mexican-American border made it more difficult to use the underground railroad. Because of this, the community died in the early 1900s. Smuggling Various smugglers also continued the underground railroad. Jim Bennett, along with Lee Quong, Louis Greenwaldt, and B.C. Springstein, all ran this business of getting Chinese immigrants into the United States for over three years. Smugglers would first collect $50 to $200 from Chinese immigrants and help them travel to the United States. The business continued successfully thanks to Bennett’s precision in ensuring that his assistants were capable and trustworthy. They would use several concealment methods, such as hiding immigrants in wagons or making the immigrants wear Mexican clothes. The business ended after Chinese inspector Charles Connell found immigrants with faulty certificates that they had received from Springstein. After capturing Springstein and hearing his testimony, Greenwaldt and Bennett were arrested, and Quong, the only Chinese smuggler of the bunch, was deported to China. It is close to impossible to determine how many Chinese immigrants entered the United States through Mexico with the help of these advanced smuggling businesses - Guamas business man W. Iberry estimated 20,000, though this could be incorrect. By 1916, the Chinese smuggling business had significantly declined, so that Chinese immigrants began to search for opportunities to live and work in Mexico. Notes Sources Category:Chinese emigration
</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>
1993–94 Portsmouth F.C. season During the 1993–94 season, the English football club Portsmouth F.C. were placed 17th out of 24 in the Football League First Division, winning 15 matches, drawing 13 and losing 18. The team reached the quarterfinals of the League Cup and the third round of the FA Cup, in both cases being knocked out after replays. First Division FA Cup League Cup Anglo-Italian Cup Category:Portsmouth F.C. seasons Portsmouth
</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>
Jag Janani Maa Vaishno Devi - Kahani Mata Rani Ki Jag Janani Maa Vaishno Devi is an Indian Hindi language mythological television series, which premiered on 30 September 2019 on Star Bharat. The series is produced by Rashmi Sharma and Pawan Kumar Marut under her company Rashmi Sharma Telefilms. Plot The story of Vaishno Devi begins when the Devas and the demons were fighting and the demons could not find anything in the Amrit Manthan and the Devas attacked the demons, then the demons settled on the earth and started making a noise there. Vaishnavi is a god in Hinduism and is portrayed as child. She tells about Parvati's ninth avatars as Jagrata. Cast Main Puja Banerjee as Vaishnavi/Matarani/Jyotawali/Sherawali/Trikuta Maisha Dixit as Young Vaishnavi/Trikuta /Matarani /Jyotawali/Sherawali Recurring Toral Rasputra as Samridhi Devi, Vaishnavi's mother Hrishikesh Pandey as Ratnakar Sagar, Vaishnavi's father Madirakshi Mundle as Lakshmi Ishita Ganguly as Parvati-Shiva's consort and Kartikeya, Ashoksundari and Ganesha's mother. Shailaputri-Daksh's daughter as Sati and later on Shailaputri as Himavat and Menavati's daughter and Parvati's first avatar. Brahmacharini-Parvati's second avatar and she wants to marry Shiva so she do tapsya for years. Chandraghanta-Parvati's third avatar and Shiva takes form Chandraskekhar to marry Parvati and she is known as Chandraganta. Kushmanda-Parvati's fourth avatar and she is light of the earth and she created Lakshmi and Saraswati. Skandamata-Parvati's fifth avatar and she is Kartikeya's mother. Katyayani-Parvati's sixth avatar and Parvati is born to sage Katyayana. (Katyayani is played by Puja Banerjee) Kaalratri-Parvati's seventh avatar and Parvati's krodh as Mahakali, Kali, Bhadrakali form. Mahagauri-Parvati's eighth avatar and sage Katyayana performs yagna so sage is born to kill the demons. Siddhidatri -Parvati's ninth avatar and she is for meditative ability and leaves her krodh. Manisha Rawat as Saraswati Vikas Salgotra as Vishnu Kapil Arya as Shiva /Mahadev-Parvati's consort and Kartikeya, Ashoksundari and Ganesha's father. Virbhadra / Mahakaal / Bhadrakaal-Mahadev's krodh avataar. Himanshu Soni as Shri Ram, Sita’s consort and King of Ayodhya Sailesh Gulabani as Indra Athar Siddiqui as Bali Preetika Chauhan as Bhudevi Preet Kaur Madhan as Kadika, wife of Svarna Sagar Vijay Badlani as Svarna Sagar, brother of Ratnakar Sagar Aarav Chowdhary as Mahipal, Senapati of Shripuram Kunal Bakshi as Vimarg Sen Vikas Verma as Asura Moor Sikandar Kharbanda as Phisachiraj Sumedh Mudgalkar as Narrator (voiceover) References Category:2019 Indian television series debuts Category:2010s Indian television series Category:Hindi-language television programs Category:Indian mythological television series Category:Indian television soap operas Category:Star Bharat television series External links
</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 Parker Robert Parker may refer to: Sir Robert Parker, 1st Baronet (c. 1655–1691), English politician; Member of Parliament for Hastings, 1679–1685 Robert Parker, Baron Parker of Waddington (1857–1918), British law lord Robert Parker (dancer), British dancer, principal dancer at Birmingham Royal Ballet Robert Parker (historian) (born 1950), Wykeham Professor of Ancient History at Oxford University Robert Parker (judge) (1796–1865), lawyer, judge and politician in New Brunswick Robert Parker (minister) (1564–1614), English Puritan scholar and divine Robert Parker (music producer), electronic musician from Stockholm, Sweden Robert Parker (musician) (1847–1937), New Zealand organist, choirmaster and conductor Robert Parker (singer) (1930–2020), American R&B singer Robert Parker (sound engineer) (1936–2004), Australian sound engineer and broadcaster Robert Parker (water polo) (born 1987), British water polo player Robert A. Parker (born 1936), astronaut Robert B. Parker (1932–2010), author of the Spenser detective novels Robert C. Parker (born 1957), United States Coast Guard officer Bob Parker (accounting scholar) (Robert Henry Parker, 1932–2016), British accounting scholar R. Hunt Parker (1892–1969), American jurist Robert Ladislav Parker, American geophysicist and mathematician Robert LeRoy Parker (1866–1908?), birth name of Butch Cassidy Robert M. Parker Jr. (born 1947), American wine critic Robert Manley Parker (born 1937), American judge Robert Townley Parker (1793–1879), British Member of Parliament for Preston Robert W. Parker (born 1960), American composer Robert W. Parker (general), U.S. Air Force general Col. Robert Parker, ring name of wrestler Robert Fuller Johnson v. Parker defendant, employer of John Casor See also Rob Parker (disambiguation) Bobby Parker (disambiguation) Bob Parker (disambiguation) Bert Parker (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>
Otto IV, Holy Roman Emperor Otto IV (1175 – 19 May 1218) was one of two rival kings of Germany from 1198 on, sole king from 1208 on, and Holy Roman Emperor from 1209 until he was forced to abdicate in 1215. The only German king of the Welf dynasty, he incurred the wrath of Pope Innocent III and was excommunicated in 1210. Career Early life Otto was the third son of Henry the Lion, Duke of Bavaria and Duke of Saxony, by his wife and Matilda of England, the eldest daughter of King Henry II of England and Eleanor of Aquitaine. His exact birthplace is not given by any original source. He grew up in England in the care of his grandfather King Henry II. Otto was fluent in French as well as German. He became the foster son of his maternal uncle, King Richard I of England. In 1190, after he left England to join the Third Crusade, Richard appointed Otto as Earl of York. The authenticity (or authority) of this grant was doubted by the vassals of Yorkshire, who prevented Otto taking possession of his earldom. Still, he probably visited Yorkshire in 1191, and he continued to claim the revenues of the earldom after becoming king of Germany, although he never secured them. Neither did he succeed in getting the 25,000 silver marks willed to him by his uncle in 1199. In 1195, Richard began negotiations to marry Otto to Margaret of Scotland, daughter and heir presumptive of King William the Lion of Scotland. Lothian, as Margaret's dowry, would be handed over to Richard for safekeeping and the counties of Northumberland and Cumberland (Carlisle) would be granted to Otto and turned over to the king of Scotland. The negotiations dragged on until August 1198, when the birth of a son and heir to William rendered them unnecessary. Having failed in his efforts to secure Otto an English earldom or else a Scottish kingdom, in September 1196 Richard, as duke of Aquitaine, enfeoffed Otto with the county of Poitou. There is some disagreement over whether Otto received Poitou in exchange for or in addition to the earldom of York. Otto was in Poitou from September 1196 until mid-1197, when he joined Richard in Normandy to confer over the appointment of bishops to the vacant sees of Poitiers, Limoges and Périgueux. He then participated in the war against Philip II of France on the side of Richard. In October he returned to Poitou. The German historian Jens Ahlers, taking into account Otto's life prior to 1198, considers that he might have been the first foreign king of Germany. Conflict with Philip of Swabia After the death of Emperor Henry VI, the majority of the princes of the Empire, situated in the south, elected Henry's brother, Philip, Duke of Swabia, king in March 1198, after receiving money and promises from Philip in exchange for their support. Those princes opposed to the Staufen dynasty also decided, on the initiative of Richard of England, to elect instead a member of the House of Welf. Otto's elder brother, Henry, was on a crusade at the time, and so the choice fell to Otto. Otto, soon recognized throughout the northwest and the lower Rhine region, was elected king by his partisans in Cologne on 9 June 1198. Otto took control of Aachen, the place of coronation, and was crowned by Adolf, Archbishop of Cologne, on 12 July 1198. This was of great symbolic importance, since the Archbishop of Cologne alone could crown the King of the Romans. The coronation was done with fake regalia,
</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>
because the actual materials were in the hands of the Staufen. Otto's election pulled the empire into the conflict between England and France. Philip had allied himself with the French king, Philip II, while Otto was supported at first by Richard I, and after his death in 1199 by his brother John. The papacy meanwhile, under Innocent III, determined to prevent the continued unification of Sicily and the Holy Roman Empire under one monarch seized the opportunity to extend its influence. Therefore, Innocent III favoured Otto, whose family had always been opposed to the house of Hohenstaufen. Otto himself also seemed willing to grant any demands that Innocent would make. The confusion in the empire allowed Innocent to drive out the imperial feudal lords from Ancona, Spoleto, and Perugia, who had been installed by Emperor Henry VI. At the same time, Innocent encouraged the cities in Tuscany to form a league, called the League of San Genesio, against imperial interests in Italy. The cities placed themselves under Innocent's protection. In 1201, Innocent announced that he recognized Otto as the only legitimate king. In return, Otto promised to support the pope's interests in Italy. Otto also had the support of Ottokar I of Bohemia, who although at first siding with Philip of Swabia, eventually threw in his lot with Otto. Otto's cause was further strengthened by the support of Valdemar II of Denmark. Philip achieved a great deal of success in the civil war that followed, allowing him in 1204 to be again crowned king, this time by the archbishop of Cologne. In the following years, Otto's situation worsened because after England's defeat by France he lost England's financial support. Many of his allies changed sides to Philip, including his brother Henry. Otto was defeated and wounded in battle by Philip on 27 July 1206, near Wassenberg, and as a consequence he also lost the support of the pope, who began to favour the apparent winner in the conflict. Otto was forced to retire to his possessions near Brunswick, leaving Philip virtually uncontested as German king. Innocent III forced the two warring parties into negotiations at Cologne, and in exchange for renouncing his claim to the throne, Philip promised Otto the hand of his daughter Beatrix in marriage, together with the Duchy of Swabia and an enormous dowry. Otto refused, and as the civil war was again about to recommence, Philip was murdered on 21 June 1208. After Philip's death, Otto made amends with the Staufen party and became engaged to Philip's daughter Beatrix. In an election in Frankfurt on 11 November 1208, he gained the support of all the electoral princes, as he promised he would not make hereditary claims to the imperial crown on behalf of any children he might father. Now fully reconciled with Innocent, Otto made preparations to be crowned Holy Roman Emperor. To secure Innocent's support, he promised to restore to the Papal States all territory that it had possessed under Louis the Pious, including the March of Ancona, the Duchy of Spoleto, the former Exarchate of Ravenna, and the Pentapolis. Travelling down via Verona, Modena, and Bologna, he eventually arrived at Milan where he received the Iron Crown of Lombardy and the title of King of Italy in 1208. He was met at Viterbo by Pope Innocent and was taken to St. Peter's Basilica, where he was crowned emperor by Pope Innocent on 21 October 1209, before rioting broke out in Rome, forcing Otto to abandon the city. Conflict with Innocent III Not content with his successes so far, Innocent also obtained from Otto further
</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>
written concessions to the Papal See, including to allow all elections of German bishops to be conducted according to Church ordinances, and not to prevent any appeals to Rome. He also promised to hand over to the Church all income from any vacant sees which had been flowing into the imperial treasury. After abandoning Rome, Otto marched north, reaching Pisa by 20 November. Here, probably advised by Peter of Celano and Dipold, Count of Acerra, he was convinced to abandon his earlier promises. Otto immediately worked to restore imperial power in Italy. After his consecration by the pope, he promised to restore the lands bequeathed to the church by the countess Matilda of Tuscany nearly a century before, and not to move against Frederick, King of Sicily. He quickly broke all his promises. He threw out the papal troops from Ancona and Spoleto, reclaiming the territory as imperial fiefs. He then demanded that Frederick of Sicily do homage for the duchies of Calabria and Apulia, and when Frederick refused to appear, Otto declared those fiefs forfeited. Otto then marched on Rome, and commanded Innocent to annul the Concordat of Worms, and to recognise the imperial crown's right to make nominations to all vacant benefices. Such actions infuriated Innocent, and Otto was promptly excommunicated by the pope for this on 18 November 1210. Subsequently, he tried to conquer Sicily, which was held by the Staufen king Frederick, under the guardianship of Innocent III. Parallel to this, the German nobility by this time were growing ever more frustrated with Otto. They felt that instead of wasting his time in Italy, and playing power politics with the pope, it was his first duty to defend the northern provinces of the empire against Valdemar II of Denmark, who had taken advantage of Otto's distractions by invading the northern provinces of the empire and possessing the whole Baltic coast from Holstein to Livonia. So while Otto was in southern Italy, several princes of the empire, including the archbishops of Mainz and Magdeburg, at the instigation of King Philip II of France and with the consent of the pope, elected Frederick King of the Romans at the Diet of Nuremberg in 1211. Otto's ambassadors from Milan appeared before the Fourth Lateran Council, pleading his case for his excommunication to be lifted. Although he claimed he had repented for his offences, and declared his willingness to be obedient to the Pope in all things, Innocent III had already recognised Frederick as emperor-elect. Otto returned to Germany to deal with the situation, hopeful to salvage something from the looming disaster. He found most of the German princes and bishops had turned against him, and that Frederick, who had made his way up the Italian peninsula, had avoided Otto's men who were guarding the passes through the Alps and had arrived at Constance. Otto soon discovered that after Beatrix died in the summer of 1212, and Frederick arrived in Germany with his army in September 1212, most of the former Staufen supporters deserted Otto for Frederick, forcing Otto to withdraw to Cologne. On 5 December 1212, Frederick was elected king for a second time by a majority of the princes. The support that Philip II of France was giving to Frederick forced King John of England to throw his weight behind his nephew Otto. The destruction of the French fleet in 1213 by the English saw John begin preparations for an invasion of France, and Otto saw a way of both destroying Frederick's French support as well as bolstering his own prestige. He agreed to join John in 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>
invasion, and in February 1214, as John advanced from the Loire, Otto was supposed to make a simultaneous attack from Flanders, together with the Count of Flanders. Unfortunately, the three armies could not coordinate their efforts effectively. It was not until John, who had been disappointed in his hope for an easy victory after being driven from Roche-au-Moine and had retreated to his transports, that the Imperial Army, with Otto at its head, assembled in the Low Countries. On 27 July 1214, the opposing armies suddenly discovered they were in close proximity to each other, on the banks of the little river Marque (a tributary of the river Deûle), near the Bridge of Bouvines. Philip's army numbered some 15,000, while the allied forces possessed around 25,000 troops, and the armies clashed at the Battle of Bouvines. It was a tight battle, but it was lost when Otto was carried off the field by his wounded and terrified horse, causing his forces to abandon the field. It is said that Philip II had sent to Frederick the imperial eagle which Otto had left lying on the battlefield. This defeat allowed Frederick to take Aachen and Cologne, as Otto was forced again to withdraw to his private possessions around Brunswick, and he was deposed in 1215. Absolved from his excommunication, he died of disease, at Harzburg castle on 19 May 1218, requesting that he be mortally expiated in atonement of his sins. Historian Kantorowicz described the death as gruesome: "deposed, dethroned, he was flung full length on the ground by the Abbot, confessing his sins, while the reluctant priests beat him bloodily to death with rods. Such was the end of the first and last Welf Emperor." He is entombed in Brunswick Cathedral. Family Otto was related to every other King of Germany. He married twice: 1209 or 1212 to Beatrice of Swabia, daughter of the German King Philip of Swabia and Irene Angelina. 19 May 1214, in Aachen to Maria of Brabant, daughter of Henry I, Duke of Brabant, and Matilda of Boulogne. Neither marriage produced any children. Notes References Abulafia, David, The New Cambridge Medieval History, Vol. V: c. 1198 – c. 1300, Cambridge University Press, 1999 Bryce, James, The Holy Roman Empire, 1913 Comyn, Robert. History of the Western Empire, from its Restoration by Charlemagne to the Accession of Charles V, Vol. I. 1851 Dunham, S. A., A History of the Germanic Empire, Vol. I, 1835 External links Literature about Otto IV in the OPAC of the Regesta Imperii Kaiserstadt Braunschweig – Otto IV Deed by Otto IV for Salem Abbey, 14 July 1209, Testament of Emperor Otto IV dated 1218 at kulturerbe.niedersachsen.de |- Category:1175 births Category:1218 deaths Category:13th-century Holy Roman Emperors Category:12th-century kings of Germany Category:13th-century kings of Germany Category:People temporarily excommunicated by the Catholic Church Category:House of Welf Category:German kings of Burgundy Category:12th-century monarchs in Europe
</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 Cumming (footballer, born 1900) David Douglas Cumming (born 1900) was a Scottish professional footballer who played as a Winger. References Category:1900 births Category:Footballers from Glasgow Category:Scottish footballers Category:Association football wingers Category:Dundee F.C. players Category:Grimsby Town F.C. players Category:English Football League players Category:Year of death 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>
Jojan Thomas Colonel Jojan Thomas was an Indian military officer with the Jat Regiment and later the 45 Rashtriya Rifles unit. He was a native of Kuttoor in Kerala and lived in Bangalore. He was commissioned from the Officers Training Academy, Chennai in March 1986. On 22 August 2008, he led an operation against terrorists in the forests of Macchal sector in the Kupwara district of Jammu and Kashmir. In the process, he killed three terrorists, but sustained several gunshot wounds and succumbed to injuries. For his bravery, he was posthumously awarded the Ashoka Chakra, the highest peace time military decoration in India. References Category:2008 deaths Category:Recipients of the Ashoka Chakra (military decoration) Category:1965 births
</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>
Time (Måns Zelmerlöw album) Time is the eighth studio album from Swedish singer Måns Zelmerlöw. It was released on 18 October 2019 through Warner Music Sweden. The album includes the singles "Walk with Me" and "Better Now". The album peaked at number eighteen on the Swedish Albums Chart. Background In August 2019, Zelmerlöw announced that he would release his new album in October. He also announced a four date tour across Sweden in October in connection with the release of the album. He said in an interview, "I am so incredibly tagged in coming back to Sweden and getting to play at home again! It is always best to play at home in Sweden and in the fall you will see that something completely new." Some recording sessions took place during songwriting camps in Battersea with others at Peter Gabriel’s Real World studios. When talking about the album, he said, "It begins with a letter from myself to my younger self as a young boy, then it moves to quite a dark place, before things turn around for the better again." Commercial performance On 25 October 2019, the album entered the Swedish Albums Chart at number eighteen, making it Zelmerlöw's first album not to debut in the top 10 in Sweden. Critical reception Scandipop gave the album a mixed review stating, "He's calling it by far his best album to date. But given the fact that earlier this year he did the unforgivable and publicly slated his earlier work, it's now somewhat difficult to invest too much into what he puts his name to, without wondering if in ten years' time he'll be referring to what you like of his right now, as pretty damn awful. Nonetheless, we'll take him for his word on this, and join him in his enthusiasm for this new record. It's funny that the track 'Time' ends up being the instrumental interlude on this album, though – as the album 'TIME' feels like an interlude itself, in his career. An album in which he stops, takes stock of the past, and readjusts to the changes in his life right now. As a result, what we end up being really excited about at the end of 'TIME', is exactly what might be coming next!" Florian Rahn from Wiwibloggs gave the album a positive review stating, "TIME is a record filled with different vibes and instruments, but it still very much sounds like the Måns Zelmerlöw we fell in love with during Eurovision 2015. He knows his sound and how to deliver a well-produced body of work." Singles "Walk with Me" was released as the lead single from the album on 1 March 2019. Måns and Dotter performed the song live during Melodifestivalen 2019. The song peaked at number 51 on the Swedish Singles Chart. "Better Now" was released as the second single from the album on 17 May 2019. The song peaked at number 79 on the Swedish Singles Chart. "One" was released as the third single from the album on 21 November 2019. The song did not enter the Swedish Singles Chart, but peaked at number 20 on the Sweden Heatseeker Songs. Track listing Charts Release history References Category:2019 albums Category:Måns Zelmerlöw albums Category:Warner Music Sweden albums
</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>
Taba Taba or TABA may refer to: Places Burundi Taba, Bururi Taba, Songa Egypt Taba, Egypt, a town in Egypt near the Gulf of Aqaba Taba Border Crossing, a border crossing between Taba in Egypt and Eilat in Israel Taba Summit, a series of negotiations between Israel and the Palestinian Authority held in Taba, Egypt in 2001 Rwanda Taba, Rwanda, a commune in Rwanda Solomon Islands Taba, Solomon Islands, a village in Guadalcanal, Solomon Islands Others Languages Taba language, an Austronesian language spoken in the northern Maluku Islands of Indonesia Others The Taba Game, a game with Greek origins popular among Argentinian gauchos TABA – Transportes Aéreos Bandeirantes, a Brazilian airline from 1945 to 1950 TABA – Transportes Aéreos da Bacia Amazônica, a Brazilian airline from 1976 to 1999
</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>
Love Songs (Bee Gees album) Love Songs is the third Bee Gees compilation album in four years, though the first to cover a specific musical style. A proposed album of love songs was in the works around 1995 when the Bee Gees recorded their own versions of "Heartbreaker" and "Emotion", but that project was soon shelved and those recordings remained unavailable until the release of Their Greatest Hits: The Record in 2001. Following the success of the Number Ones compilation in 2004, Universal once again tried to mine the Bee Gees catalog, this time focusing on their ballads. Spanning their entire career, Love Songs features many of the group's big hits, but also includes some lesser known tracks such as "Secret Love" and "For Whom the Bell Tolls", which were both big hits in Europe. Also included is a live version of "Islands in the Stream". The U.S. and the UK versions differed slightly in song selection and running time. Included in the UK version were 1993's "Heart Like Mine" and a song Barry and Maurice wrote together with Ronan Keating in 1999 called "Lovers and Friends", backing Keating both vocally and instrumentally while also producing the recording. Another anomaly was the inclusion of the Robin Gibb solo hit "Juliet" from 1983, which was a big hit in many countries, except in the UK and the U.S. By the time Love Songs was released, there had been several Bee Gees compilations on the market as well as their entire back catalog, so this release seemed somewhat redundant. In the U.S. it did chart but only managed to reach #166. In their homeland it climbed to No. 51 in the UK Albums Chart, but its best showing was in France where it went Top 20, peaking at No. 18. Track listing USA/International "To Love Somebody" (Bee Gees' 1st) "Words" (Horizontal) "First of May" (Odessa) "Lonely Days" (2 Years On) "How Can You Mend a Broken Heart" (Trafalgar) "How Deep Is Your Love" (Saturday Night Fever soundtrack) "More Than a Woman" (Saturday Night Fever soundtrack) "(Our Love) Don't Throw It All Away" (Bee Gees Greatest) "Emotion" (Their Greatest Hits: The Record) "Too Much Heaven" (Spirits Having Flown) "Heartbreaker" (Their Greatest Hits: The Record) "Islands in the Stream" (Live) (One Night Only) "Juliet" (How Old Are You?) - Robin Gibb solo track) "Secret Love" (High Civilization) "For Whom the Bell Tolls" (Size Isn't Everything) "Closer Than Close" (Still Waters) "I Could Not Love You More" (Still Waters) "Wedding Day" (This Is Where I Came In) UK/Japan "To Love Somebody" "Words" "First of May" "Lonely Days" "How Can You Mend a Broken Heart" "How Deep Is Your Love" "More Than a Woman" "(Our Love) Don't Throw It All Away" "Emotion" "Too Much Heaven" "Heartbreaker" "Islands in the Stream (Live)" "Juliet" "Secret Love" "For Whom the Bell Tolls" "Heart Like Mine" (Bonus track) (Size Isn't Everything) "Closer Than Close" "I Could Not Love You More" "Wedding Day" "Lovers and Friends" (feat. Ronan Keating) (Bonus track) - 5:17 (New song) PRC "To Love Somebody" "Words" "First of May" "Lonely Days" "How Can You Mend a Broken Heart" "How Deep Is Your Love" "More Than a Woman" "(Our Love) Don't Throw It All Away" "Emotion" "Too Much Heaven" "Heartbreaker" "Juliet" "Secret Love" "For Whom the Bell Tolls" "Closer Than Close" "I Could Not Love You More" "Wedding Day" References Category:Bee Gees compilation albums Category:2005 compilation albums Category:Universal Records compilation albums Category:Albums recorded at IBC Studios
</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>
Lattarico Lattarico () is a town and comune in the province of Cosenza in the Calabria region of southern Italy. References Category:Cities and towns in Calabria
</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>
Russell Corner Historic District The Russell Corner Historic District of Princeton, Massachusetts encompasses a small 19th century commercial center that developed into a summer resort area by the early 20th century. Centered at the intersection of Merriam Road and Gregory Hill Road, the district includes fifteen contributing buildings on , which are predominantly residential in nature, and were mostly built before 1830. Some of the properties are found on East Princeton and Sterling Roads, and Bullock Lane. The district was listed on the National Register of Historic Places in 2006. Both Merriam Road and Gregory Hill Road have been major transportation routes through Princeton; Merriam Road in particular was laid out in part along a Native American path in the mid-17th century. The oldest building in the district is the c. 1748 house of Abijah Wood at 16 Merriam Road, a Georgian house with a late-19th-century wraparound porch. The second house was built c. 1760, probably by Peter Goodnow; portions of this house survive in the building at 49 Gregory Hill Road. John Russell moved to the area in 1787, and soon afterward opened a tavern and inn. His son Charles was an energetic farmer and shopowner who was also active in the town and church; it is for him the area is named. In part through the economic activity of the Russells, a small cluster of houses and businesses arose. The village came to include a schoolhouse in 1810 (now at 43 Gregory Hill Road), and by 1830 eight buildings stood near the intersection, all but one of which still stand. The village was largely bypassed by the resort hotel developments that grew in other parts of Princeton, but smaller scale development of summer residences, including conversion of some of the older buildings, took place in Russell's Corner. See also National Register of Historic Places listings in Worcester County, Massachusetts References Category:Buildings and structures in Princeton, Massachusetts Category:Historic districts in Worcester County, Massachusetts Category:Historic districts on the National Register of Historic Places in Massachusetts Category:National Register of Historic Places in Worcester County, Massachusetts
</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 Sydney Mystery The Sydney Mystery is a point-and-click Adventure game released by American studio Twilight Software in April 13, 2003. The game takes place in Sydney, Australia. Plot In the city of Sydney in Australia, a retired detective has been kidnapped. The detective's niece looks for leads to rescue him. Gameplay The gameplay is simplified without the need to click on any verbal commands or items in an inventory. The player interacts with the current area by hovering the mouse over particular objects or any people present and left clicking to perform an action. The mouse pointer changes to an icon that matches the appropriate action. Right clicking cycles through any available items the player possess. Left clicking while holding the selected item uses it. Whenever the player leaves a starting area, the game switches to an over world map of Sydney, allowing the player to travel to any known locations to progress. Development Brendan Reville was inspired by many story-driven Sierra and LucasArts adventure games to create his own, and distanced his title from other FMV video games instead hoping to make a first-person adventure game that happened to be photographed and filmed on location. He started the creation of the game from scratch in the year 2000. He programmed the entire game engine AGE (Adventure Game Engine), but much work had the project in progress for three years. In addition to coding the engine, Reville almost singlehandedly designed the game, photographed and filmed all live action sequences, scripted the game, compiled and created the art, video and sound, programmed the puzzles, recorded the dialogue, tested the game, made a demo, created an installer, built the website, and did PR. The game's budget was around $500. By Monday 9 September 2002, Brendan had presented his game to the Independent Games Festival. Filming Filming of the game was done in the very plot-centered city Sydney itself around 17 different locations with over 240 scenes in total. Capturing pictures was done using a wide-angled SLR Camera provided by Shanno Sanders. Paul Fiore also lent a digital video camera which was used together with a Shotgun microphone for quality filming. Due to the low-budget, the cast of 11 actors were either friends or friends-of-friends of Reville, and only one cast member had previous acting experience. Reception Michelle Whicker of PC Gameworld argued that the game could be used as a public relations marketing tool for Sydney due to its use of several scenic locations throughout the city. Meanwhile Alex Tait of Just Adventure described the game as the equivalent of buying a CD based on hearing one song on the radio only to be disappointed, and thought its unintentional humour was stylistically to that found in The Curse of Monkey Island and The Feeble Files. References External links Category:2003 video games Category:Adventure games Category:Detective video games Category:Video games developed in the United States Category:Video games featuring female protagonists Category:Video games set in Australia Category:Windows games Category:Windows-only games Category:Point-and-click adventure games
</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>
Abu Ratib Mohamed Mustapha Ali Masfaka (born 1962), better known as Abu Ratib (), is a Syrian-born Nasheed singer of Islamic and Arabic music based on classical Arab poetry. Early life Masfaka was born in Aleppo, Syria. From an early age, he was immersed in musical traditions of the Arab world. At a young age, he joined the Arab Music Institute in Aleppo, Syria where he studied musical notation (Solfège) and learned to play the violin. Masfaka studied broadcasting and television direction in Amman, Jordan, and obtained experience in the broadcasting engineering. Career As Masfaka balanced music with a growing interest in Islam, his work moved in a spiritual direction. He joined the band of the singer Abi Dujanah for religious singing, until he founded Al-Huda band in 1981 and began issuing the albums of Islamic singing. In 1986, Al-Huda band developed to found Al-Huda Islamic Art. In 1991, he founded Al-Huda International Art Production. Masfaka is an active member of the Arabic arts community and is involved in numerous organizations to promote Islamic music and artist development worldwide. He has traveled over the U.S and abroad as a featured artist at community events in mosques and on college campuses. He is a member of Michigan-based Crescent Academy International, and aims to overcome negative impressions of Islam through his works such as a play about the Islam featuring the Islamic prophet Muhammad which attracted a large non-Muslim audience. He is also head of the World Islamic Art League. Masfaka has often sung in support of human rights for the occupied Palestinian people. Awards In 2006, Masfaka was awarded the World Youth Prize for Islamic Work in the field of Inshad (religious singing). Personal life Masfaka is married to Batoul Qaweqji and has five daughters, Ayah, Nour, Sana, Hiba, and Sama Masfaka. See also Syrian people Music of Syria References External links Category:1962 births Category:Living people Category:Date of birth missing (living people) Category:Syrian Muslims Category:Syrian emigrants to the United States Category:Syrian male singers Category:Performers of Islamic music Category:People from Aleppo Category:Salahaddin University-Erbil alumni Category:University of Jordan alumni Category:Mohammed V University alumni
</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>
Band-e Sheram Band-e Sheram is a village in Balkh Province in northern Afghanistan. See also Balkh Province References External links Satellite map at Maplandia.com Category:Populated places in Balkh Province
</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>
Minorka Mercado Minorka Marisela Mercado Carrero (born January 15, 1972 in Caracas, Venezuela) is Miss Venezuela 1993. She won the Miss Photogenic, Best in Swimsuit, Best in Maria Clara gown awards and placed second runner-up to Miss Universe 1994. Pageantry Miss Venezuela Minorka competed in 1993 as Miss Apure in her country's national beauty pageant, Miss Venezuela, capturing the crown and the right to represent her country in Miss Universe 1994, where she finished as 2nd runner up. She also represented Venezuela at Miss International 1995 in Tokyo, Japan but did not place. Miss Universe As the official representative of her country to the 1994 Miss Universe pageant held in Manila, Philippines on May 20, 1994, she won three special awards: Best in Swimsuit, Best in Filipino Costume and Miss Photogenic. In the end, she placed second runner-up to Miss Universe 1994. References External links Miss Venezuela Official Website Miss Universe Official Website Category:1972 births Category:Living people Category:Miss Universe 1994 contestants Category:Miss Venezuela winners Category:People from Caracas Category:Venezuelan women's volleyball 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>
Chase Strumpf Chase Aaron Strumpf (born March 8, 1998) is an American professional baseball second baseman in the Chicago Cubs organization. Career Strumpf attended JSerra Catholic High School in San Juan Capistrano, California. During his freshman year, he committed to play college baseball at UCLA. After his freshman year, he played on the USA Baseball 15U National Team, helping Team USA win a gold medal in Barranquilla, Colombia. He was not drafted out of high school in the 2016 MLB draft and thus enrolled at UCLA. In 2017, Strumpf's freshman year, he appeared in 55 games (making 54 starts), hitting .239 with seven home runs and thirty RBIs. That summer, he played in the Northwoods League where he was named an All-Star after batting .335 with six home runs. As a sophomore at UCLA in 2018, Strumpf started 58 games in which he slashed .363/.475/.633 with 12 home runs and 53 RBIs and was named to the Pac-12 First Team. After the season, he was selected to play for the USA Baseball Collegiate National Team, but was unable to participate due to injury. Prior to his 2019 junior season, Strumpf was named a preseason All-American by numerous media outlets including D1Baseball and Perfect Game. He was named to the Pac-12 First Team for the second consecutive year, and finished his junior year batting .279 with nine home runs and 44 RBIs in 63 games. Strumpf was considered one of the top prospects for the 2019 MLB draft. He was selected by the Chicago Cubs with the 64th overall pick, and signed for $1.05 million. He made his professional debut with the Arizona League Cubs and was promoted to the Eugene Emeralds after seven games. After 26 games with Eugene, he was promoted to the South Bend Cubs, with whom he finished the year. Over 39 games between the three clubs, Strumpf slashed .244/.374/.400 with three home runs and 17 RBIs. References External links UCLA Bruins bio Category:1998 births Category:Living people Category:Baseball players from California Category:Baseball second basemen Category:UCLA Bruins baseball players Category:Arizona League Cubs players Category:Eugene Emeralds players Category:South Bend Cubs 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>
European Film Award for Best Actor The European Film Award for Best Actor is an award given out at the annual European Film Awards to recognize an actor who has delivered an outstanding leading performance in a film industry. The awards are presented by the European Film Academy (EFA) and was first presented in 1988. Winners and nominees 1980s 1990s 2000s 2010s Berlin Film Festival Silver Bear Award for Best Actor Cannes Film Festival Award for Best Actor Venice Film Festival Volpi Cup for Best Actor Most wins by actor See also BAFTA Award for Best Actor in a Leading Role BIFA Award for Best Performance by an Actor in a British Independent Film César Award for Best Actor David di Donatello for Best Actor Goya Award for Best Actor Polish Academy Award for Best Actor Robert Award for Best Actor in a Leading Role External links European Film Academy archive Actor Category:Film awards for lead actor Category:Awards established in 1988 Category:1988 establishments in Europe
</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>
Czerwony Folwark Czerwony Folwark is a village in the administrative district of Gmina Suwałki, within Suwałki County, Podlaskie Voivodeship, in north-eastern Poland. References Czerwony Folwark
</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>
Shantel Bailey Shantel Suleon Bailey (born 30 April 1995) is a Jamaican footballer who plays as a midfielder for the Jamaica women's national team. International career Bailey made her senior debut for Jamaica on 20 April 2018 at the 2018 CFU Women's Challenge Series. International goals Scores and results list Jamaica's goal tally first References Category:1995 births Category:Living people Category:Women's association football midfielders Category:Jamaican women's footballers Category:Jamaica women's international footballers
</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>
Palmerston North Palmerston North (; ) is a city in the North Island of New Zealand and the seat of the Manawatū-Whanganui region. Located in the eastern Manawatu Plains, the city is near the north bank of the Manawatu River, from the river's mouth, and from the end of the Manawatu Gorge, about north of the capital, Wellington. Palmerston North is the country's seventh-largest city and eighth-largest urban area, with an urban population of The official limits of the city take in rural areas to the south, north-east, north-west and west of the main urban area, extending to the Tararua Ranges; including the town of Ashhurst at the mouth of the Manawatu Gorge, the villages of Bunnythorpe and Longburn in the north and west respectively. The city covers a land area of . The city's location was once little more than a clearing in a forest and occupied by small communities of Māori, who called it Papa-i-Oea, believed to mean "How beautiful it is". In the mid-19th century, it was settled by Europeans—originally by Scandinavians and, later, British colonists. On foundation, the British settlement was bestowed the name Palmerston, in honour of Viscount Palmerston, a former British Prime Minister. The suffix North was added in 1871 to distinguish the settlement from Palmerston in the South Island. Today, the name is often informally shortened to "Palmy". Early Palmerston North relied on public works and sawmilling. The west coast railway was built in 1886, linking the town to Wellington, and Palmerston North benefited from a booming pastoral farming industry. Linton Military Camp, Palmerston North Hospital, and the establishment of Massey University (in 1927) have reduced the dependence on farming since the early 20th century. Popular attractions include Te Manawa (a museum and art gallery that includes the New Zealand Rugby Museum), and several performing arts venues. History Early settlement Ngāti Rangitāne were the local Māori iwi (tangata whenua) living in the area known as Te Ahu-ā-Tūranga, when a trader, Jack Duff, became the earliest known European to explore the area . He came on a whaling ship and explored possibly as far inland as the site of Woodville. He reported his discovery on arrival back to Porirua. Colonel Wakefield heard of the potential that the Manawatu had for development and visited in 1840. In 1846 Charles Hartley, another trader, heard from tangata whenua of a clearing in the Papaioea forest and he proceeded through the dense bush and forest and discovered it for Europeans. In 1858, the Government began negotiations with local iwi to purchase land in Manawatu. There was a dispute at the time between rival iwi Ngāti Rangitāne and Ngāti Raukawa as to who has the right to sell. The dispute is resolved in favour of Rangitāne. On a visit in 1859, John Tiffin Stewart, an employee of the Wellington Provincial Council, was shown the Papaioea clearing by Rangitāne chief, Te Hirawanu, and noted its suitability for a "good site for a township". In 1864, Te Ahu-a-Turanga Block was sold by Rangitāne to the Government for £12,000, in an effort to open the Manawatu to settlement. Stewart returned in 1866 on behalf of the Wellington Provincial Council (under whose jurisdiction the new purchase fell) and made the original survey and subdivision in the Papaioea forest clearing. The settlement, named Palmerston to commemorate the recently deceased Prime Minister of Great Britain, was laid out according to Stewart's plan consisting of a series of wide and straight streets in a rectangular pattern. The focal point was an open space of subsequently known as The Square. On 3 October 1866, Palmerston was formally
</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>
endorsed after Isaac Earl Featherston (Wellington Provincial Superintendent) signed a proclamation defining the boundaries of the settlement. The first sections were sold after. Among the first settlers included Scandinavians, who arrived in 1871. They established settlements at Awapuni and Whakarongo/Stoney Creek. Later the same year, the suffix North was added to distinguish the settlement of the same name in Otago. In 1872 a petition was launched to change the name of the settlement. A public meeting in 1873 ends with no clear decision on the name. The railway line was laid through the Square in 1875. The foundation stone for the original All Saints Church was laid by Louisa Snelson on 29 September 1875. By 1875 there were newspapers, a doctor and a post office. In 1876, Palmerston North became a Local Board District, within the Wellington Provincial Council. This existed until the abolition of the provinces later the same year. Also in the same year, the council set aside land north of the Manawatu River for the purposes of a reserve. In 1890, this land was again set aside and would become in 1897, the Victoria Esplanade. Growing population By 1877, when the Borough Council came into existence, Palmerston North was an isolated village in the midst of the native forest that covered inland Manawatu. By 1878, the population was approximately 800 people and sawmilling was the main industry of the district. The arrival of the railway in 1886 saw an increase in the speed of growth and the town was at the centre of a lucrative agricultural district. The opening of the nearby Longburn Freezing Works provided employment, while the Borough Council instigated more infrastructural schemes such as the sewerage system. The Railway through the Manawatu Gorge to Napier was completed in 1891. In 1893, Rangitāne sold the Hokowhitu block, increasing the area of land available for settlement. In the same year, the Public Hospital opened in a wooden building on Terrace Street (now Ruahine Street). The hospital required significant fundraising. At the end of the decade, the Boer War broke out in South Africa and men from Palmerston North were among the volunteers. By 1900 the population had reached 6,000. In the 1910s Palmerston North's growth was steady. The population in 1911 about 10,991 (excluding Māori). The city was affected by World War I, with Awapuni Racecourse being used as an army training camp in 1914. During the course of the war, the Borough Council renamed all German-sounding and foreign street names. When the war finished in 1918, celebrations were delayed due to the Influenza epidemic. City status In 1930, the population reached the 20,000 threshold and Palmerston North was officially proclaimed a city, the 7th in New Zealand. Development was slow due to the great depression and World War II. An airport was established at Milson in 1936, which is now Palmerston North Airport. From 1938, the first Labour government (1935–1949) initiated state housing programmes in West End (Savage Crescent precinct) and Roslyn. In 1941, the Manawatu River flooded again, having last flooded in 1902. Large parts of Hokowhitu and Awapuni were underwater, with residents evacuating to higher ground (peaked at 5.8m). In 1940 the Māori Battalion was formed in Palmerston North and trained at the Showgrounds (now Arena Manawatu). In 1942, Linton Army Camp was established. After the war, the city's growth was rapid. By 1950, the city's boundaries had extended to include Milson and Kelvin Grove. In 1953, the boundaries would further extend to include Awapuni, which in the same year, was again flooded by the Manawatu River, along with Hokowhitu. It was 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>
largest flood since 1902. Although work had started in 1926, it was not until 1959 the Milson Deviation of the North Island Main Trunk was opened. This meant future trains would pass to the north of the city, instead of through the Square. Later in 1963, the railway station at Tremaine Avenue opens. The last trains passed through the Square in 1964. In 1961, the Highbury was added to the council area. In 1963, Massey University College of Manawatu was formed by the amalgamation of the Massey Agricultural College with the Palmerston North University College. In 1964, it becomes Massey University, an autonomous tertiary learning institution with the power to grant its own degrees. In 1967, city boundaries were again extended to include land in Aokautere, Kelvin Grove, Milson, Amberley (Westbrook) and Awapuni as part of future growth for the next 25 years. In 1970, the New Zealand Rugby Museum was established and a tribute to the founding father of rugby in New Zealand, Charles Munro, was opened at Massey University. In 1971, a competition to design a civic building for the vacant railway land at the Square, is won by Wellington architects, Maurice and John Patience. The resulting building was finished in 1979. In 1976, the Manawatu and Oroua rivers flood, 24-hour rainfall records in Feilding and Palmerston North are exceeded and some residents from both locations are evacuated. In 1977, Palmerston North City Council celebrated its Centenary of Municipal government. The Queen and the Duke of Edinburgh are among visitors to Palmerston North. On 1 November 1989, New Zealand local government authorities were reorganised. Palmerston North City boundaries were extended to include Ashhurst, Linton and Turitea through amalgamation of parts of the former Kairanga County, Oroua County and Ashhurst Town Council. On 1 July 2012, Bunnythorpe, Longburn, part of the area around Kairanga and an area around Ashhurst were joined to Palmerston North City Council area. Geography Although the land Palmerston North is situated on is bounded by the lofty Ruahine and Tararua ranges in the east and south respectively, the city has a predominantly flat appearance. The occasional rise in elevation occurs further away from the river and is especially pronounced in the north and northeast, and also on the south side of the river. The typical urban area elevation ranges between 20–40 metres (65–130 ft) above sea level. The highest point is above sea level. This is in the Tararua ranges, south-east of Scotts Road. The lowest point is above sea level. This is at the river bank near Te Puna Road. Incidentally, both these locations are in the south-west of the city, by Linton. There are dedicated to public reserves. The length of the Manawatu river within the city boundary is and its tributary at Ashhurst, the Pohangina, is . Climate Palmerston North's climate is temperate with warm summer afternoon temperatures of 20 – in summer and in winter. On average temperatures rise above on 20 days of the year. Annual rainfall is approximately with rain occurring approximately 5% of the time. There are on average 200 rain-free days each year. In the ranges that flank the city there is often sustained wind, especially in spring. Much of this land is within the city boundaries and these ranges have the reputation of providing the most consistent wind in the country. Close to the city is the largest electricity-generating series of wind farms in the southern hemisphere, with 286 turbines in the Tararua and Ruahine Ranges providing power for approximately 50,000 homes. Governance Palmerston North City Council Palmerston North is governed by a city council consisting
</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>
of the mayor and 15 councillors, elected on a citywide basis. New Zealand's local government elections occur simultaneously nationwide every three years, with the next election to take place in late 2019. Before the 2016 election councillors were elected representing one of five city wards, but Palmerston North's electoral wards were abolished by the Local Government Commission in 2013. The 2013 election also saw the single transferable vote (STV) system introduced, replacing first past the post (FPP). In 2017 the Council voted to create a Māori ward, but the decision was overturned by a city-wide referendum in May 2018. As of 2019 there are 11 independent councillors, two affiliated with the Green Party and two with Labour. Current council (2019–2022): Asterisk denotes Deputy Mayor The Council's functions are broken down into six units: Strategy & Planning Customer Finance Community Infrastructure Marketing & Communications The Chief Executive is Heather Shotter, who replaces Paddy Clifford. Horizons (Manawatū-Whanganui) Regional Council Palmerston North is the seat of the council. For electoral and regional representation purposes, Palmerston North City makes up the Palmerston North constituency and is represented by 4 councillors: Palmerston North Constituency Councillors as at 2019 election: Wiremu Te Awe Awe Jono Naylor Fiona Gordon Rachel Keedwell Central governance Since the 2017 general election, Palmerston North has been represented by three electorate MPs and one List MP in Parliament. Palmerston North The boundaries of the electorate of Palmerston North have changed often since the introduction of MMP. It has been previously larger than now and formerly extended over the Manawatu River into the southern suburbs. A 2007 redraw transferred the southern suburbs to Rangitīkei, in return for Kelvin Grove and Milson, making Palmerston North a completely urban electorate. Currently Palmerston North is surrounded by the Rangitīkei electorate. Since 2008, the Palmerston North electorate has been held by Iain Lees-Galloway of Labour. Rangitīkei The large electorate of Rangitīkei extends south to Palmerston North's southern and eastern parts, the mostly rural former Ashhurst-Fitzherbert ward. Rangitīkei therefore completely surrounds the electorate of Palmerston North (the only electorate thus surrounded completely by another). The electorate was formerly represented (1998–2011) by Simon Power (National) and has been represented by former Manawatu District Mayor Ian McKelvie (National) since the New Zealand general election at the end of that year. Te Tai Hauāuru One of the seven Māori electorates, Te Tai Hauāuru includes all of the Palmerston North area. The electorate is currently held by Adrian Rurawhe of the Labour Party. List members MP Darroch Ball is on the New Zealand First party list. Justice The Palmerston North Courthouse in Main Street (east) is a combined District and High Court and serves the city and surrounding area. Demographics At the 2013 census, Palmerston North had a population of 80,079, an increase of 2,355 people (3.0%) since the 2006 census. There were 29,892 occupied dwellings, 1,914 unoccupied dwellings, and 99 dwellings under construction. Of the population, 38,391 (47.9%) were male and 41,688 (52.1%) female. The city had a median age of 33.8 years, 4.2 years below the national median age of 38.0 years. People aged 65 years and over made up 13.3% of the population, compared to 14.3% nationally, and people under 15 years made up 20.0% of the population, compared to 20.4% nationally. Due to Palmerston North being a university city, approximately 36% of the population is aged between 15.0 and 24.9 years. The median annual income of all people 15 years and over was $27,000, compared with $28,500 nationally. Of those, 39.9% earned under $20,000, compared with 38.2% nationally, while 24.1% earned over $50,000, compared to 26.7% nationally. Palmerston
</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>
North City had an unemployment rate of 7.5% of people 15 years and over, compared to 7.1% nationally. Culture and identity At the 2013 Census of Population and Dwellings, 78.9 percent of Palmerston North residents identified as of European ethnicity, 16.5 percent as Māori, 9.7 percent as Asian, 4.5 percent as Pacific Peoples, 1.3 percent as Middle Eastern/Latin American/African, and 2.1 percent as another ethnicity (mainly 'New Zealander'). Around 19.2 percent of the Palmerston North's population was born overseas. Asia is the largest region of origin, accounting for 36.9 percent of the overseas-born population in the region. Religion At the 2013 Census, around 48.9 percent of Palmerston North residents affiliate with Christianity and 5.4 percent affiliate with non-Christian religions, while 43.4 percent are irreligious. Anglicanism and Catholicism are the largest Christian denominations in Palmerston North with 12.1 percent and 12.0 percent or residents affiliating respectively. Hinduism (1.5 percent), Islam (1.3 percent), and Buddhism (1.3 percent) are the largest non-Christian religions. Palmerston North is a cathedral city, and the seat of the Roman Catholic Diocese of Palmerston North. The Cathedral of the Holy Spirit, Palmerston North is its cathedral. The Bishop of Palmerston North is Charles Drennan. In the Anglican Communion, Palmerston North is under the jurisdiction of the Diocese of Wellington, under Bishop Justin Duckworth. Palmerston North is also in the Anglican Hui Amorangi of Te Pīhopatanga o Te Upoko o Te Ika, under current Pīhopa Rev. Muru Walters. There are also many other churches with denominations such as Adventist (Mosaic Community Church and Palmerston North Seventh Day Adventist Church), Apostolic, Assembly of God (AOG), Baptist, Brethren, Christian Scientist, Church of Christ, Church of Jesus Christ of Latter-day Saints, Jehovah's Witnesses, Lutheran, Methodist, Pentecostal, Presbyterian and Religious Society of Friends. There are Sikh gurdwara near the CBD and in Awapuni. There is an Islamic centre in West End, and an Islamic prayer centre at Massey University. Economy Although Palmerston North has 1.9% of the population of New Zealand, 2.5% of the nation's employees work in the city. The important sectors are tertiary education, research and central government (Defence force). At the 2013 census, the largest employment industries for Palmerston North residents were health care and social assistance (4,686 people, 13.0%), education and training (4,473 people, 12.4%), retail trade (4,062 people, 11.2%), public administration and safety (3,636 people, 10.1%), and manufacturing (2,703 people, 7.5%). The tertiary education sector provides NZ$500 million a year to the local economy and the education sector accounted for 11.3% of the Palmerston North workforce in February 2006. Palmerston North has economic strengths in research, especially in the bio-industry, defence, distribution and smart business sectors. The city is home to more than 70 major educational and research institutions, including New Zealand's fastest expanding university, Massey University; the Massey University Sport and Recreation Institute at the Massey University campus, Turitea; Universal College of Learning (UCOL) and Linton Army Camp. Business innovation Palmerston North has a long history of innovation in the business sector. A number of firms founded in or near the city have become nationally or internationally renowned. One of the largest and well known of these is GSK (GlaxoSmithKline). Glaxo had its beginnings in Bunnythorpe, now a part of Palmerston North. Companies with National Head Offices based in Palmerston North: Toyota New Zealand Higgins Group FMG Insurance New Zealand Pharmaceuticals Limited Plumbing World Limited Steelfort Engineering Limited Hino Distributors (NZ) Ltd Hunting and Fishing New Zealand CB Norwood Distributors Ltd Sime Darby Commercial (NZ) Ltd Motor Truck Distributors (NZ) Ltd (National distributors of Mack, Renault and Volvo Trucks & Buses) Truck Stops (NZ) Ltd OBO 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>
facilities and attractions Palmerston North has a number of facilities and attractions. It is also the gateway to attractions in other parts of the region, such as Tongariro National Park, Ruahine and Tararua Ranges. When Palmerston North Airport serviced international flights, the city was also an international gateway to Hawke's Bay, Whanganui and Taranaki. The Plaza The Plaza is the largest shopping centre in the Manawatū-Whanganui region and a key shopping centre in the lower North Island, boasting over 100 stores. It was built in about the 1980s and then upgraded again in around the late 2000s to what it is now. Downtown Downtown on Broadway combines retail and boutique shopping and Downtown Cinemas. The Square Originally part of the Papaioea clearing, The Square is a seven-hectare park of lawn, trees, lakes, fountains, and gardens in the centre of the city. It is the city's original park and also the centrepoint from whence the city's main streets are arranged. The Square contains the city's war memorial and a memorial dedicated to Te Peeti Te Awe Awe, the Rangitāne chief instrumental in the sale of Palmerston North district to the government in 1865. Near the centre of the park is the Clock Tower with its illuminated cross and coloured lights. Also here is the city's iSite, the Civic Building (seat of the City Council), the City Library, Square Edge and the commercial heart of Palmerston North's CBD. Retail stores (including the Plaza) and eateries line the road surrounding the park. In around 1878, a Māori contingent, including Te Awe Awe, gathered together to choose a Māori name for The Square. They chose Te Marae o Hine, meaning "The Courtyard of the Daughter of Peace". This name reflected their hope all people of all races would live together in enduring peace. Parks and recreational facilities Palmerston North and its surroundings feature roughly 100 parks and reserves. Most notable is Victoria Esplanade, a park located along the northern bank of the Manawatu River west of Fitzherbert Avenue. It was opened in 1897 to commemorate the 60th Jubilee of Queen Victoria's reign, and includes a native bush reserve along the river bank, formal botanical gardens, playgrounds and the Café Esplanade, all connected with walking and bicycle tracks. Located within the Esplanade are several attractions. The Peter Black Conservatory is a large tropical greenhouse built in 1941 and refurbished in 2014. The Wildbase Recovery visitor centre and aviary for recovering wildlife opened in 2019 and is run in partnership with Massey University's Veterinary Hospital. The volunteer-operated Esplanade Scenic Railway features a miniature railway track offering 20 minute rides through native bush. New Zealand's largest rose garden, the Dugald McKenzie Rose Garden, is the site of the New Zealand international rose trials and contributed to the city's one-time nickname, "Rose City". It was recognised in 2003 by the World Federation of Rose Societies as one of the finest rose gardens in the world. Adjacent to the Esplanade are the multi-sport playing fields of Ongley and Manawaroa Parks, the twin turf hockey fields and Fitzherbert Park, the premier cricket ground. Elsewhere in the city are parks for sports like rugby, such as Coronation Park, Bill Brown Park and Colquhoun Park (also used for softball/baseball); and football: Skoglund Park (home of the Central Football Federation) and Celaeno Park. Memorial Park has a football pitch, and also has family-friendly facilities. The Hokowhitu Lagoon is also located nearby the Esplanade and is a popular site for recreational kayaking and canoeing. The Lido Aquatic Centre is Palmerston North's largest aquatic centre. Freyberg Community Pool is an all-year indoor swimming pool complex
</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>
located next to Freyberg High School in Roslyn. Riverbank development The city council since 2012 has been beautifying the banks of the Manawatu River, opening up previously neglected areas into more accessible recreational parks and reserves. Part of this development is He Ara Kotahi, a pedestrian and cycle track that connects the city to Massey University, adjacent research institutes and Linton Military Camp, all located on the south side of the Manawatu River. The opening in 2019 included the opening of Palmerston North's second bridge, a pedestrian bridge connecting the Holiday Park on Dittmer Drive to the track across the river. As well as a recreational asset, He Ara Kotahi was designed to be a pedestrian and cycle commuter route between the city bridge and Massey University, to improve safety and relieve vehicle traffic volumes on the Fitzherbert Bridge. The remainder of the track to Linton crosses tributary streams with boardwalks and four smaller bridges. Libraries Palmerston North has a main public library with five branches and one mobile library. The central Palmerston North Library is located in the Square and houses the main collections. The other four are located in Ashhurst, Awapuni, Linton, Roslyn and Te Pātikitiki (Highbury). Youth Space Near the Square is the Youth Space, opened in September 2011 as a dedicated place for Palmerston North's many young people to congregate in a safe environment. Youth space is free to all, and provides table-tennis, gaming consoles, musical instruments, library books, iPads, a kitchen, and other services. Culture Arts Te Manawa is the cultural museum of art, science and history. Attached to Te Manawa is the New Zealand Rugby Museum. There are many small independent galleries. Many of New Zealand's best-known artists came from or live in Palmerston North. The list includes Rita Angus, John Bevan Ford, Shane Cotton, Paul Dibble, Pat Hanly, Brent Harris, Bob Jahnke, John Panting, Carl Sydow and Tim Wilson. Performing arts Palmerston North houses multiple theatres which regularly host musical performances, theatrical plays and formal events. These theatres include Regent on Broadway Theatre is a 1393-seat multipurpose performing arts facility. Centrepoint Theatre is a prominent professional theatre and the only one outside the main centres of New Zealand. Globe Theatre is a small community theatre of around 200 seats, opened in November 1982 as a partnership between the City Council and the Manawatu Theatre Society: both parties contributed capital funds to enable the building to be constructed. The building was designed by local architect, Brian Elliot, who returned to design the Theatre’s major redevelopments in 2014, with the addition of a second auditorium and an extension to the foyer and cafe/bar area. The redevelopment won a New Zealand Architecture Award for Elliot's work. Top comics including John Clarke, Jon Bridges, Jeremy Corbett, Tom Scott all come from Palmerston North as do stage, television and film performers Paul Barrett, Shane Cortese, Kate Louise Elliott, Simon Ferry, Greg Johnson, Jeff Kingsford-Brown and Alison Quigan. Music Palmerston North has a thriving musical scene with many national and international acts touring through the town, and many local acts performing regularly. Local groups include the Manawatu Sinfonia and Manawatu Youth Orchestra (MYO) who perform throughout the year. The Manawatu Youth Orchestra celebrated its 50th year in September 2011. Sport Participation in sport is an important and popular pastime in Palmerston North. Representation at a national level is predominantly provincial based, meaning most sports teams representing Palmerston North also draw their players from other towns from around Manawatu. The premier multi-sports venue in Palmerston North is Arena Manawatu, which is known as the Central Energy Trust Arena after 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>
headline sponsors. The main stadium (Arena One) is the home of the Manawatu Turbos rugby union team and the Robertson Holden International Speedway. There are indoor venues at Arena Manawatu where netball, basketball, volleyball and Badminton are played. Other important venues include Memorial Park, Fitzherbert Park, Celaeno Park, Manawaroa/Ongley Park, Skoglund Park, Vautier Park and Massey University sports fields. # : not based in Manawatu, however, home ground when playing in Manawatu. Infrastructure and services Health Palmerston North Hospital is the city's main public hospital, and is the seat of the MidCentral District Health Board. The hospital is the major trauma centre for Palmerston North and the Manawatu, Horowhenua and Tararua districts. There were two private hospitals, Aorangi and Southern Cross. In 2012 these two private surgical hospitals merged and are now known as Crest Hospital. Electricity The Palmerston North Municipal Electricity Department (MED) was formed in 1924 to supply the city with electricity. The Manawatu-Oroua Electric Power Board (EPB) supplied the surrounding rural areas. Electricity was initially generated at the Keith Street power station until the transmission lines from Mangahao Power Station to Bunnythorpe substation were completed. The Keith Street power station continued to regularly generate electricity until the Inter-Island HVDC link was commissioned in 1965, when it was relegated to standby duty before finally being decommissioned in 1992. The Palmerston North MED and Mawawatu Oroua EPB were dissolved in the late 1990s with the government electricity sector reforms. The retail business was sold to Genesis Energy while the lines business became part of Powerco. Today, Powerco continues to operate the local distribution network, with electricity fed from two Transpower substations, at Bunnythorpe and at Linton. There is now a competitive electricity retail market, although Genesis continues to be the dominant retailer in the city. Three wind farms are located on the Ruahine and Tararua Ranges bordering Palmerston North: Te Apiti windfarm, Tararua windfarm and Te Rere Hau Wind Farm. The three farms together have a maximum output of 300 MW. Natural gas Palmerston North was one of the original nine towns and cities in New Zealand to be supplied with natural gas when the Kapuni gas field entered production in 1970 and a 260 km high pressure pipeline from Kapuni south to Wellington (including a 27 km lateral pipeline from Himatangi to supply Palmerston North) was completed. The high pressure transmission pipelines supplying the city are now owned and operated by First Gas, with Powerco owning and operating the medium and low pressure distribution pipelines within the city. At the 2013 census, 40.5% of Palmerston North homes were heated wholly or partially by natural gas, the highest in New Zealand. Internet and telephone Fibre to the premises is being deployed in Palmerston North as part of the Government's Ultra-Fast Broadband programme. , the fibre roll-out in the city is 94 percent complete, with a 42.2 percent uptake rate. Transport Palmerston North is a significant road and rail junction. As such, it is an important distribution hub for the Central and lower North Island, with many freight distribution centres based here. Road The city's main roads are all in dual carriageway format and arranged in a grid pattern, with four streets of significance radiating from The Square in the CBD. They are: State highways Palmerston North is served by four state highways: State Highway 3 runs northwest–southeast from SH 1 at Sanson through central Palmerston North to SH 2 at Woodville. The section from Sanson forms the main route from the upper North Island, Taranaki and Whanganui into Palmerston North, while the section from Woodville forms the main route
</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>
from the Hawke's Bay into Palmerston North. State Highway 57 runs southwest–northeast from SH 1 at Ohau, south of Levin, through the southern outskirts of Palmerston North to SH 3 east of Ashhurst. It forms the main route from Wellington to southern and eastern Palmerston North. State Highway 56 runs southwest–northeast from SH 57 at Makeura, northeast of Shannon, New Zealand, to the intersection of Pioneer Highway and Maxwells Line in the suburb of Awapuni. It forms the main route from Wellington to northern and western Palmerston North. State Highway 54 runs north-south from SH 1 at Vinegar Hill, north of Hunterville, through Feilding to SH 3 at Newbury, on the northern Palmerston North border. It provides an alternative route from the Upper North island into Palmerston North Cycling Palmerston North is perceived as being better for cycling than most New Zealand cities, with 2001 figures putting it a close second only to Blenheim in terms of bicycle modal share. However, by 2006, cycling to work had almost halved in a decade to 5.4% and the 2013 census found that, in the central city, only 6 cycled, but 690 travelled by motor vehicle. The Manawatu River Pathway is great for family or beginners riders, as it is all flat to mildly contoured, with some limestone sections, as well as wide cement paths. The track has many access points to this trail, which runs for over 9 km between Maxwells Line in the West to Riverside Drive in the East. A new 3 km section has been added between Ashhurst and Raukawa Road, with plans to link this to the existing path over the next two years, making over 22 km of scenic tracks to explore alongside the river. Palmerston North has a fairly comprehensive 65 km on-road bicycle lane network, particularly in high traffic areas, to make it safer for people to get around the city by bike. All local buses have racks for two cycles. The cycle lane network has been criticised for a number of reasons. Motor traffic is often too fast, and there is no physical barrier between bicyclists and motorists. Most bicycle lanes in the city are marked out with parking spaces for motorist parking, making those lanes 'pointless' and raising the risk of motorists opening car doors into the path of passing bicyclists. Rebecca Oaten, the so-called 'Helmet Lady' who campaigned nationwide in the late 1980s for a New Zealand bicycle helmet law, is from Palmerston North. Bus Urban services are coordinated by Horizons Regional Council, through Masterton-based bus company, Tranzit. Seven urban buses leave the terminal in Main Street East (in front of Palmerston North Courthouse) at least every half-hour. The buses are assigned to loop routes servicing different parts of the city. Daily services run to the nearby towns of Linton, Ashhurst, Feilding (via Airport), Foxton, Levin and Marton. Inter-regional routes are operated by Intercity and Tranzit. Intercity's routes run south (to Wellington), north (towards Auckland, via Whanganui, Rotorua and Taupo, or Napier) and east (to Masterton) from the bus terminal in The Square, Palmerston North. Air Palmerston North Airport is located in the suburb of Milson, approximately north of the central business district. It is a regional gateway to the central North Island region. The airport has regular services to domestic destinations including Auckland, Christchurch, Hamilton, Nelson and Wellington. Flights are currently served by Air New Zealand Link and Originair. The airport is presently the operational base of the Massey University School of Aviation. The airport is also an freight hub for Parcelair. Rail Palmerston North is an important passenger and freight
</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>
stop on the North Island Main Trunk Railway. There are only two passenger trains run by KiwiRail: the weekday-only Capital Connection commuter train once a day to and from Wellington, and is a stop for the Northern Explorer to and from Auckland and Wellington. Until 1964, the railway ran through the city centre, with Palmerston North Railway Station in The Square. The station was moved and the track diverted to the north by the Milson Deviation in 1959–1963; work on the deviation had started in 1926. Near the current railway station, the North Island Main Trunk railway is joined by the Palmerston North - Gisborne Line, which runs through the Manawatu Gorge to Woodville and Hawke's Bay. A connection to the Wairarapa Line is at Woodville. Education Palmerston North is considered "the student city" of New Zealand North Island. It is an important base for tertiary institutions. The home campuses of Massey University, Universal College of Learning (UCOL) and Institute of the Pacific United (IPU) are here. A large proportion of Palmerston North's population consists of students attending these institutions or the various Papaioea Campuses of Te Wananga o Aotearoa during the student year. Tertiary Education Institutions Primary and Secondary Schools Palmerston North has five state secondary schools: Palmerston North Boys' High School and Queen Elizabeth College in the north, Freyberg High School in the north-east, Palmerston North Girls High School in the south, and Awatapu College in the south-west. The city also has one state-integrated Catholic secondary school, St Peter's College. Palmerston North also has one special character secondary school: Cornerstone Christian School, Palmerston North Media The major daily newspaper is the Manawatu Standard. The weekly community newspaper is the Guardian. Most radio stations are based outside the city, however some on the nationwide networks have studios in Palmerston North. Local radio stations include More FM (formerly "2XS") featuring the popular "Mike West in the Morning" breakfast show from 6am-10am. Sister station The Breeze also has a local breakfast show from 6am-10am hosted by Burnzee. Radio Control is the local alternative student radio station, featuring local personality Abi Symes on "The Continental Breakfast" from 7 am to 9 am. Access Manawatu 999AM is a local community station and Kia Ora FM is the local Iwi station. The city's main television and FM radio transmitter is located atop Wharite Peak, northeast of the city centre. The first transmitter at the site was commissioned in 1963 to relay Wellington's WNTV1 channel (now part of TVNZ 1). Sister cities Palmerston North has four sister cities: Missoula, Montana, United States Guiyang, People's Republic of China Kunshan, People's Republic of China Mihara, Hiroshima, Japan Personalities Grant Robertson (born 1971), Labour MP; born in Palmerston North. Current MP for Wellington Central. Shane Cortese (born 1968), actor and singer Joseph Nathan (1835–1912), founder of Glaxo (since merged to become the multinational GlaxoSmithKline). Alan Loveday (1928–2016), violinist Ross Taylor (born 1984), former New Zealand Black Caps captain and Central Districts batsman (cricket) Matthew Conger (born 1978), current FIFA international football referee Brendon Hartley (born 1989), racing driver and World Endurance Championship champion Named after the city An NAC Vickers Viscount (ZK-NAI) was named "City of Palmerston North". This aircraft was withdrawn from NAC service in 1975. An Ansett New Zealand de Havilland Canada Dash 8 (ZK-NES) was also named "City of Palmerston North", staying in service until the airline's demise. References External links Official websites Official Website of Palmerston North City & Manawatu Palmerston North City Council Street Maps and Property Information Unofficial portals The Square Palmerston North portal Palmerston North City Portal Category:Main urban areas 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>
New Zealand Category:University towns in New Zealand
</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>
Monte Lindo River The Monte Lindo River is a river of Paraguay. See also List of rivers of Paraguay References Rand McNally, The New International Atlas, 1993. GEOnet Names Server Category:Rivers of Paraguay
</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>
Aruba at the 2011 Pan American Games Aruba competed at the 2011 Pan American Games in Guadalajara, Mexico from October 14 to 30, 2011. Aruba competed with seventeen athletes in five sports. Bowling Aruba had qualified a women's team of 2 athletes. Women Individual Pairs Cycling BMX Aruba had qualified one male BMX cyclist. Men Swimming Aruba had qualified three swimmers. Men Women Synchronized swimming Aruba had qualified a full team (a duet and a team). 9 athletes. Taekwondo Aruba had qualified a team of 2 male taekwondo athletes. Men References Category:Nations at the 2011 Pan American Games P 2011
</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>
Eddie Aleong Eddie Aleong (born 4 May 1937) is a Trinidadian former cricketer. He played four first-class matches for Trinidad and Tobago between 1959/60 and 1965/66. References External links Category:1937 births Category:Living people Category:Trinidad and Tobago 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>
1976 Campeonato Paulista The 1976 Campeonato Paulista da Divisão Especial de Futebol Profissional was the 75th season of São Paulo's top professional football league. Palmeiras won the championship by the 18th time. no teams were relegated. Championship The championship was divided into two rounds; in the first, the eighteen teams of the championship were divided into three groups of six teams, with each team playing once against all other teams, and the four best teams of each group passing to the Second round. The team with the most points in the first phase regardless of group would gain a bonus point for the Second round. In the second round, the remaining twelve teams would all play against each other once, and the team with the most points would be champion. Although no teams were relegated from last year's championship, Saad, which had been invited into the last two championships, wasn't invited back into the championship for 1976. First round Group A Group B Group C Second round References Category:Campeonato Paulista seasons Paulista
</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>
Sandra Hess Sandra Hess (born March 27, 1968) is a Swiss actress and fashion model. She is best known for her role as Sonya Blade in the film Mortal Kombat: Annihilation, and Lieutenant Alexandra 'Ice' Jensen in the television series Pensacola: Wings of Gold. Life and career Hess began modeling and working on television commercials when she was 15. After completing high school, she entered the University of Zurich to study law but before completing her degree she came to the United States to build an acting career. Once settled in Los Angeles, Hess started taking acting classes. Her first role was in the 1992 film Encino Man, playing a cave-woman to Brendan Fraser's caveman character. In 1997, she played the role of Sonya Blade in the film Mortal Kombat: Annihilation, taking over the role played by Bridgette Wilson in the first film. In 1998, she portrayed Immortal bounty hunter Reagan Cole, a friend of Duncan MacLeod's, in the ninth episode of the sixth season of Highlander: The Series. She also portrayed Andrea von Strucker / Viper in the TV movie Nick Fury: Agent of S.H.I.E.L.D., and Alexandra "Ice" Jensen in Pensacola: Wings of Gold. Her guest spots include the series Lois & Clark: The New Adventures of Superman, Sliders (Season 4 – Genesis), SeaQuest DSV and CSI: Crime Scene Investigation. In 2008, Hess starred as Sasha in General Hospital. In 2010, she guested in Psych. She married actor Michael Trucco in July 2009; they were wed in Mexico. She and Trucco both played the parts of Lieutenants in the same Marine unit in the Pensacola: Wings of Gold series. Filmography External links Category:1968 births Category:Living people Category:Swiss film actresses Category:Swiss television actresses Category:20th-century Swiss actresses Category:21st-century Swiss actresses Category:People from Zürich
</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>
John Turner (Archdeacon of Taunton) John Turner (c.1734–1817) was an English churchman, Archdeacon of Taunton from 19 September 1780 until his death on 28 March 1817. He matriculated at Hertford College, Oxford in 1751, aged 17, graduating B.A. in 1755. Notes Category:18th-century Anglican priests Category:Archdeacons of Taunton Category:1817 deaths Category:Alumni of Hertford College, Oxford Category:Year of birth uncertain
</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>
Randy Rogers Band (album) Randy Rogers Band is the sixth album released by the Randy Rogers Band, an American country music group. This is their second album on a major label. "In My Arms Instead" was the first single released from the album; it entered the Billboard Hot Country Songs chart at #58. Track listing "Wicked Ways" (Jon Richardson) - 4:20 "Better Than I Ought To Be" (Gary Nicholson, Randy Rogers) - 2:51 "Lonely Too Long" (George Ducas, Rogers) - 3:19 "One Woman" (Rogers, Stephony Smith) - 4:05 "Never Be That High" (Rogers, Smith) - 3:42 "Didn't Know You Could" (Micky Braun, Rogers) - 3:58 "In My Arms Instead" (Sean McConnell, Rogers) 5:18 "When The Circus Leaves Town" (Clint Ingersoll, Jon Richardson, Rogers) - 4:22 "Buy Myself A Chance" (McConnell, Rogers) - 3:41 "Break Even" (Geoffrey Hill) - 4:58 "Let It Go" (Radney Foster, Rogers) - 3:32 "This Is Goodbye" (Ingersoll, Heather Morgan) - 3:25 Personnel Randy Rogers Band Brady Black - fiddle, background vocals Geoffrey Hill - electric guitar, background vocals Les Lawless - drums Jon Richardson - bass guitar Randy Rogers - acoustic guitar, lead vocals Additional Musicians Jessi Alexander - background vocals Eric Borash - dobro, acoustic guitar, baritone guitar, electric guitar, lap steel guitar, mandolin Todd Cooper - background vocals Clayton E. Corn - Hammond B-3 organ, keyboards, Wurlitzer Radney Foster - acoustic guitar, electric guitar Craig Krampf - percussion Sean McConnell - background vocals Jon Randall - background vocals Chart performance References Category:2008 albums Category:Mercury Nashville albums Category:Randy Rogers Band albums
</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>
Bráz Magaldi Bráz Magaldi (born June 1897, date of death unknown) was a Brazilian sports shooter. He competed in the 25 m rapid fire pistol event at the 1932 Summer Olympics. References Category:1897 births Category:Year of death missing Category:Brazilian male sport shooters Category:Olympic shooters of Brazil Category:Shooters at the 1932 Summer Olympics Category:People from Minas Gerais
</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 law clerks of the Supreme Court of the United States (Seat 10) Law clerks have assisted the justices of the United States Supreme Court in various capacities since the first one was hired by Justice Horace Gray in 1882. Each justice is permitted to have between three and four law clerks per Court term. Most persons serving in this capacity are recent law school graduates (and typically graduated at the top of their class). Among their many functions, clerks do legal research that assists justices in deciding what cases to accept and what questions to ask during oral arguments, prepare memoranda, and draft orders and opinions. After retiring from the Court, a justice may continue to employ a law clerk, who may be assigned to provide additional assistance to an active justice or may assist the retired justice when sitting by designation with a lower court. Table of law clerks The following is a table of law clerks serving the associate justice holding Supreme Court seat 10 (the Court's tenth associate justice seat by order of creation), which was established on April 10, 1869 by the 41st Congress through the Judiciary Act of 1869 (). This seat is currently occupied by Justice Clarence Thomas. |} |} |} |} |} |} |} |} Notes References Additional sources Baier, Paul R. (1973). "The Law Clerks: Profile of an Institution," Vanderbilt L. Rev. 26: 1125–77. "Finding Aid to Thurgood Marshall Papers," Library of Congress, list of clerks. "Georgia Law Alumni Who Have Clerked for a U.S. Supreme Court Justice," Advocate, Spring/Summer 2004 (listing 6 names). Judicial Clerkship Handbook, USC Gould Law School, 2013-2014, p. 33, Appendix B. "List of law clerks," The Papers of Justice Tom C. Clark, Tarlton Law Library, University of Texas Law School. Retrieved August 11, 2016. Newland, Charles A. (June 1961). "Personal Assistants to the Supreme Court Justices: The Law Clerks," Oregon L. Rev. 40: 306–07. News of Supreme Court clerks. University of Virginia Law School, list of clerks, 2004-2018. University of Michigan clerks to the Supreme Court, 1991-2017, University of Michigan Law School Web site (2016). Retrieved September 20, 2016. Ward, Artemus and David L. Weiden (2006). Sorcerers' Apprentices: 100 Years of Law Clerks at the United States Supreme Court. New York, NY: New York University Press. , . External links Supreme Court of the United States official website SCOTUS Justices: How Do Their Clerks Help Them?, Marcia Coyle of the National Law Journal, by PBS NewsHour, via youtube Seat 10
</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>
Sapu Mhicha Sapu Mhichā (Newari language: सपू म्हिचा, "tripe bag") is a special dish consisting of buffalo leaf tripe stuffed with bone marrow. It is boiled and fried. Sapu Mhichā is a specialty of the Newari cuisine of the Kathmandu Valley and is prepared during special occasions. The delicacy is one of the dishes served to honor a man when he goes to the home of his wife's parents for festival dinners. It is served after the main course and before dessert. Production Water buffalo leaf tripe is cut into small pieces and the layers separated to form bags. Diced bone marrow is stuffed into the bags. The opening is gathered and tied with a piece of thread to close it. The bags are then boiled and fried. The eater puts the whole Sapu Mhichā in the mouth and bites it off, leaving the tied end between the forefinger and thumb. This way the melted bone marrow remains inside the mouth when it gushes out as the bag is ripped. See also List of Nepalese dishes List of stuffed dishes References Category:Offal Category:Nepalese cuisine Category:Meat dishes Category:Newari cuisine Category:Stuffed dishes
</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>
International Center, Dallas International Center is a neighborhood of high-rise office and residential buildings in the Oak Lawn area of Dallas, Texas (USA). The land in the neighborhood is owned primarily by Harwood International, a development firm based in Dallas. The area is in proximity to Victory Park, Uptown, and the Arts District of downtown, three highly popular districts in Central Dallas. Like its neighborhoods, International Center is experiencing large amounts of growth in both the commercial and residential sectors. External links Neighborhood Masterplan Emporis listing of structures
</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>
Kalateh-ye Kuchak Kalateh-ye Kuchak (, also Romanized as Kalāteh-ye Kūchak) is a village in Howmeh Rural District, in the Central District of Semnan County, Semnan Province, Iran. At the 2006 census, its existence was noted, but its population was not reported. References Category:Populated places in Semnan 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>
New Haven High School (Indiana) New Haven High School is a public secondary education school for grades nine through twelve. New Haven High School is one of five high schools in the East Allen County Schools district, Indiana. The district has adopted a unique method called “five campus”, meaning a student can take a class offered at any one of the schools in the district. History The first school in New Haven was built downtown in 1885. The first true high school was built in 1923 and was officially called the Adams-Township school, although everyone referred to it as New Haven school. It accommodated kindergarten through 12th grade. In the 1940s a separate building was put up next to this school to house the elementary grades. In 1955 and 1957 additions were made to the high school to allow for the growing student population. To further aid with the growth a middle school was built across the street on the high school’s football field and track. (Goeglein) Soon enough it was time to build a new high school. The high school moved to a piece of land across the highway. The middle school moved into the old high school and the elementary moved into the middle school. The elementary is now used for offices and has a newly renovated auditorium that is used for special performances. The current New Haven High School was built in 1975 and opened in 1977. The building is located on a property – the smallest amount of land of any school in its district. It cost approximately $7 million to build. The gym is a smaller version of the nearby Harding High School gym. The school was originally designed to accommodate 1500 students. Recently the school has had some modifications to classrooms to allow for different kinds of classes. Currently the school’s capacity is 1030. This is the first school in the district to have an elevator in it. The principal of the school when it opened was Paul Goeglein. Extracurricular activities NHHS provides numerous clubs and other activities for its students. Included in the list are: Academic Decathlon - a multi disciplinary academic competition where students compete in 10 categories (English, Music, Art, Science, etc.) of a particular subject matter (Civil War, China, Mexico, etc.) Concert Band Channel 86 - a student run news channel covering important school related events Drama Club French Club Key Club Latin Club Marching Band / Color Guard National Honor Society Peer Tutoring Pep Band S.A.D.D. - Students Against Destructive Decisions Show Choir Spanish Club Winter Guard Notable alumni Norm Ellenberger, former (retired) college basketball coach, New Mexico Lobos Matt Bischoff, MLB player for the New York Mets and the Seattle Mariners Dan Brown, player for the Chicago Cubs. Single game strikeout record holder with 19. Single season ERA 0.91. Career ERA 1.18. Both school records. 24-13 career record. Inducted into the New Haven Bulldogs Athletics Hall of Fame in 2006. See also List of high schools in Indiana Northeast Eight Conference New Haven, Indiana References External links Official Website Athletics Football History Boys Basketball History (1993–present) Girls Basketball History (2000–present) Category:Public high schools in Indiana Category:Schools in Allen County, Indiana
</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>
Mindy Sterling Mindy Lee Sterling (born July 11, 1953) is an American actress and voice actress. She portrayed Frau Farbissina in the Austin Powers series of comedy films, Principal Susan Skidmore on the Disney Channel series A.N.T. Farm, Ms. Endive in the Cartoon Network series Chowder, and on Nickelodeon as Lin Beifong in The Legend of Korra, and Miss Francine Briggs on iCarly. Early life Sterling was born in Paterson, New Jersey. Her father is actor and comedian Dick Sterling. Her mother was a dancer. She has one brother, Mark Sterling. She is Jewish. Career Early career Around 1973, Sterling landed a recurring role on the syndicated children's series Dusty's Treehouse. She later joined the L.A.-based comedy troupe The Groundlings. Film career Despite having starred in numerous films in the 1980s and 1990s, it was her character Frau Farbissina, the diminutive and domineering Germanic cohort of Dr. Evil (Mike Myers) in Austin Powers: International Man of Mystery, that brought Sterling high praise for her work in films. In 1999, following the success of Austin Powers: The Spy Who Shagged Me, Sterling appeared in Drop Dead Gorgeous (also 1999), and later reprised the character in Austin Powers in Goldmember. In 2000, she played one of the townspeople in the live-action adaptation of How the Grinch Stole Christmas, and also provided additional voices in the 2018 CGI version, The Grinch. In 2012, she appeared in a book trailer for a parody of The Hunger Games entitled The Hunger Pains. Television career From 1990-91, she was one of the comedians in the satirical show On the Television. Sterling plays Christian Slater's secretary Arlene Scott in the show My Own Worst Enemy. She also played Judge Foodie on the Disney show That's So Raven and a volleyball coach on The Suite Life of Zack & Cody. She appeared as the wedding planner on the Friends episode "The One with Barry and Mindy's Wedding" in 1996, and in 2004, as the casting director on the episode "Joey and the Big Audition" of Joey. Sterling has achieved success in teen sitcoms, playing Ms. Francine Briggs on iCarly, and in A.N.T. Farm as Principal Susan Skidmore. On April 25, 2013 before the finale of the second season of A.N.T. Farm, Sterling announced she would not be appearing on the show's third season due to family obligations. She announced in June 2012 that she may return in the fourth season, with guest appearances in season three. In 2000, Sterling appeared as a celebrity guest on Hollywood Squares. She returned to the show in 2002. Between 2002 and 2004, Sterling made guest appearances on the Donny Osmond version of the game show Pyramid. In 2010, she had a recurring role as bitter neighbour Mitzi Kinsky in Desperate Housewives. In 2013, she currently stars as Janice Nugent in the comedy series Legit. In 2018, she appeared in Netflix's A Series of Unfortunate Events. Voice acting career From 2007-2010, Sterling played Ms. Endive, the main antagonist to Mung Daal in Chowder. She also voiced the character Lin Beifong, the second Police Chief of Republic City and daughter of original chief Toph Beifong, in The Legend of Korra. Her many other voice-over credits include guest roles on The Wild Thornberrys, Invader Zim, Ice Age: The Meltdown, American Dragon: Jake Long, Higglytown Heroes, Robot Chicken, Mars Needs Moms, The Looney Tunes Show, Justice League Unlimited, Winx Club, Kick Buttowski: Suburban Daredevil and Scooby-Doo! Mystery Incorporated. Personal life Family Sterling has one son, Max, b. 1994, who began a career in stand-up comedy in May 2016. Health Sterling was diagnosed with breast cancer 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>
a routine mammogram in 1998. After a lumpectomy, chemotherapy, radiation treatment and tamoxifen, she has remained cancer-free. Filmography Film Television Video games References External links Category:1953 births Category:Actresses from New Jersey Category:American film actresses Category:American television actresses Category:American voice actresses Category:Breast cancer survivors Category:Living people Category:Actors from Paterson, New Jersey Category:20th-century American actresses Category:21st-century American actresses
</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>
Was zählt "Was zählt" (What counts) is a song by Die Toten Hosen. It's the first single and the third track from the album Auswärtsspiel. The song is an ode to love ("Wenn nur die Liebe zählt" - When only love matters), as it describes, how one can go through anything and suffer a lot for love, also give up any memories for it. Music video The music video was directed by Ralf Schmerberg. It shows the band performing the song in a room in a city building. Also the streets outside are shown, which are overflowing with police, as the video was shot a while after the September 11 attacks. Track listing "Was zählt" (What counts) (Breitkopf, von Holst/Frege) − 4:37 "Hängt ihn höher" (Hang him higher) (Meurer/Frege) - 2:37 "Drüber reden" (Talking about it) (von Holst/Frege) – 1:42 "Schöner warten" (Nicer waiting) (Frege/Frege) – 3:58 Charts Category:2001 singles Category:Die Toten Hosen songs Category:Songs written by Campino (singer) Category:Songs written by Andreas von Holst Category:Songs written by Michael Breitkopf Category:2001 songs
</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>
Australian cricket team in England in 1899 The Australian cricket team in England in 1899 played 35 first-class matches including five Tests, the first time that a series in England had consisted of more than three matches. It was also the first time that a panel of selectors was appointed; previously the authority for the ground where the match was to be played was responsible for selecting the side. The First Test at Trent Bridge saw W G Grace make his final appearance for England, while Wilfred Rhodes made his Test debut in the same match. Test series summary Australia won the Test series 1-0, with four matches drawn. First Test Second Test Third Test Fourth Test Fifth Test External sources CricketArchive – tour summaries Annual reviews James Lillywhite's Cricketers' Annual (Red Lilly) 1900 Wisden Cricketers' Almanack 1900 Further reading Bill Frindall, The Wisden Book of Test Cricket 1877-1978, Wisden, 1979 Chris Harte, A History of Australian Cricket, Andre Deutsch, 1993 Ray Robinson, On Top Down Under, Cassell, 1975 Category:1899 in Australian cricket Category:1899 in English cricket 1899 Category:English cricket seasons from 1890 to 1918 Category:International cricket competitions from 1888–89 to 1918 1899
</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>
Miss Nicaragua 2015 The Miss Nicaragua 2015 pageant, was held on March 7, 2015 in Managua, after several weeks of events. At the conclusion of the final night of competition, Daniela Torres Bonilla from Managua won the title. She represented Nicaragua at Miss Universe 2015 later that year. The rest of the finalists would enter different pageants. Placements Special awards Miss Gillette Venus - Granada - Ruth Angélica Martínez Best Hair - Managua - Daniela Torres Most Beautiful Face - Tipitapa - Yaoska Ruiz Miss Photogenic - Managua - Daniela Torres Miss Congeniality - Matagalpa - Karen Salgado Best Smile - Managua - Daniela Torres Miss Fitness - Tipitapa - Yaoska Ruiz Official Contestants Trivia Celeste Castillo Miss International Nicaragua 2013 roaming around backstage and spoke to misses for their take on whether couture has any influence on their street style or double check their answers, to report on the action for the live broadcasting Show. Viewers were able to interact with the pageant via Movistar. Fans were able to vote for their favorite contestant through the Miss Nicaragua website, from February 7 to March 7. Miss Tipitapa, Yaoska Ruiz, won the most MSM votes among the contestants, winning her a spot in the Top 6. Judges Mignone Vega - Fashion Designer Marianela Lacayo - Miss Nicaragua 2002 Claudia Salmeron - Miss Nicaragua 2003 Rafael Sanchez - Executive Director of AFN (American Nicaraguan Foundation) Vannesa Osorio - Marketing Manager of Contempo Hotel Tatiana Pilarte Arcia - Independent Consultor Indiana Sánchez - Miss Nicaragua 2009 Live Music Opening Show – Ballet Macehuatl - "El Garañon" Swimsuit Competition – Alias Rhythm - "Closer" Evening Gown Competition – Mafia Funk & Polux - "Dias De Amar", "Fantasma Camaleon" & "Yo No Se Mañana" Special Guests Grupo de Danzas S.e.v.E.n - "Uptown Funk" References Category:Miss Nicaragua Category:2015 in Nicaragua Category:2015 beauty pageants
</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>
Marinobacter xestospongiae Marinobacter xestospongiae is a Gram-negative, slightly halophilic and non-spore-forming bacterium from the genus of Marinobacter which has been isolated from the sponge Xestospongia testudinaria from the sea coast of Saudi Arabia. References External links Type strain of Marinobacter xestospongiae at BacDive - the Bacterial Diversity Metadatabase Category:Alteromonadales Category:Bacteria described in 2012 Category:Halophiles
</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>
Peter Theophil Riess Peter Theophil Riess (27 June 1804 – 22 October 1883) was a German physicist, known mostly for his work in electricity, particularly friction electricity, but also in the field of electromagnetic induction. In the latter work he developed two devices, the spark micrometer (sometimes called the Riess micrometer) and the Riess spiral coils, both of which were used by Heinrich Hertz in his experiments to prove the propagation of electromagnetic waves. Riess was the son of a wealthy jeweller in Berlin. He received his doctorate in 1831 from the University of Berlin. In the early years after graduation he stayed with his father who was ill and helped run the business. Even after his father died, he never sought an academic teaching position. However he was a fellow (Mitglied) of the Berlin Academy, the first Jewish member of that academy, and under its auspices, and within its journal, Riess published prolifically. He also published several books, as well as many articles in the Annalen der Physik. Riess was a friend of most of the leading members of the Berlin intelligentsia of the period, and had interests in history, literature, art and music. He died in Berlin at age 79. References Category:1804 births Category:1883 deaths Category:German physicists
</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>
Ice Ice Baby "Ice Ice Baby" is a hip hop song written by American rapper Vanilla Ice, K. Kennedy, and DJ Earthquake. It was based on the bassline of "Under Pressure" by Queen and David Bowie, who did not receive songwriting credit or royalties until after it had become a hit. Released on his debut album, To the Extreme, it is his best known song. It has appeared in remixed form on Platinum Underground and Vanilla Ice Is Back! A live version appears on the album Extremely Live, while a nu metal version appears on the album Hard to Swallow, under the title "Too Cold". "Ice Ice Baby" was first released as the B-side to Vanilla Ice's cover of "Play That Funky Music", but the single was not initially successful. When disc jockey David Morales played "Ice Ice Baby" instead, it began to gain success. "Ice Ice Baby" was the first hip hop single to top the Billboard Hot 100. Outside the United States, the song topped the charts in Australia, Belgium, the Netherlands, New Zealand, Ireland and the United Kingdom, thus helping the song diversify hip hop by introducing it to a mainstream audience. It was also used in Smallfoot (film) and in the first trailer for The Angry Birds Movie 2. Lyrics and music Robert Van Winkle, better known by his stage name Vanilla Ice, wrote "Ice Ice Baby" in 1983 at the age of 16, basing its lyrics upon his experiences in South Florida. The lyrics describe a shooting and Van Winkle's rhyming skills. The chorus of "Ice Ice Baby" originates from the signature chant of the national African American fraternity Alpha Phi Alpha. Of the song's lyrics, Van Winkle stated in a 2001 interview that "If you released 'Ice Ice Baby' today, it would fit in today's lyrical respect among peers, you know what I'm sayin'? [...] My lyrics aren't, 'Pump it up, go! Go!' At least I'm sayin' somethin'." The song's hook samples the bassline of the 1981 song "Under Pressure" by Queen and David Bowie, who did not receive credit or royalties for the sample. In a 1990 interview, Van Winkle claimed the two melodies were slightly different because he had added an additional note on the "and" of the fourth beat, an anacrusis ("pickup") between odd-numbered and subsequent even-numbered iterations of the "Under Pressure" sample. In later interviews, Van Winkle readily admitted he sampled the song and claimed his 1990 statement was a joke; others, however, suggested he had been serious. After representatives for Queen and Bowie threatened a copyright infringement suit against him, the matter was settled out of court, with Van Winkle being required to pay financial recompense to the original artists. Bowie and all members of Queen were also given songwriting credit for the sample. In December 1990, Van Winkle told British youth music magazine Smash Hits where he came up with the idea of sampling "Under Pressure": Van Winkle described himself as the first rapper to cross into the pop market and said that although his pioneer status forced him to "take the heat for a lot of people" for his music's use of samples, the criticism he received over sample use allowed sampling to become acceptable in mainstream hip hop. Release "Ice Ice Baby" was initially released by Ichiban Records as the B-side to Van Winkle's cover of "Play That Funky Music". The 12-inch single featured the radio, instrumental and a cappella versions of "Play That Funky Music" and the radio version and "Miami Drop" remix of "Ice Ice Baby". When a disc jockey named David Morales played
</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>
"Ice Ice Baby" instead of the single's A-side, the song gained more success than "Play That Funky Music". A music video for "Ice Ice Baby" was produced for $8000. The video was financed by Van Winkle's manager, Tommy Quon, and shot on the roof of a warehouse in Dallas, Texas. In the video, Van Winkle is shown rapping the lyrics while he and others dance to the song. Heavy airplay of the video by The Box while Van Winkle was still unknown increased public interest in the song. "Ice Ice Baby" was given its own single, released in 1990 by SBK Records in the United States, and EMI Records in the United Kingdom. The SBK single contained the "Miami Drop", instrumental and radio mixes of "Ice Ice Baby" and the album version of "It's A Party". The EMI single contained the club and radio mixes of the song, and the shortened radio edit. The single was quickly pulled from the American market soon after the song reached number one, in a successful attempt to drive consumers to buy the album instead. Reception "Ice Ice Baby" garnered critical acclaim, and was the first hip hop single to top the Billboard charts. It has been credited for helping diversify hip hop by introducing it to a mainstream audience. Entertainment Weekly reviewer Mim Udovitch wrote that "[Vanilla Ice] probably would have scored with his hit rap single 'Ice Ice Baby' even if he hadn't been white. There's just something about the way its hook – a sample from Queen and David Bowie's 'Under Pressure' — grabs you and flings you out onto the dance floor." Network 40 wrote: "Like Mellow Man Ace, the rap melts slowly and is as much a mood piece as it is a cruising tune. A motocross champion from Dallas via Miami, the 22-year-old Ice says it's time to chill out." Following the song's success, California rapper Mario "Chocolate" Johnson, an associate of record producer Suge Knight, claimed that he had helped in writing the song, and had not received credit or royalties. Knight and two bodyguards arrived at The Palm in West Hollywood, where Van Winkle was eating. After shoving Van Winkle's bodyguards aside, Knight and his own bodyguards sat down in front of Van Winkle, staring at him before finally asking "How you doin'?" Similar incidents were repeated on several occasions before Knight showed up at Van Winkle's suite on the fifteenth floor of the Bel Age Hotel, accompanied by Johnson and a member of the Los Angeles Raiders. According to Van Winkle, Knight took him out on the balcony by himself, and implied that he would throw Van Winkle off unless he signed the rights to the song over to Knight. Legacy After audiences began to view Van Winkle as a novelty act and a pop star rather than a legitimate rapper, his popularity began to decline. Detroit-based rapper Eminem states that when he first heard "Ice Ice Baby", "I felt like I didn't want to rap anymore. I was so mad, because he was making it real hard for me." Van Winkle lost some credibility among hip hop fans, but later began to regain some success, attracting a new audience outside of the mainstream audience that had formerly accepted him and then rejected him. "Ice Ice Baby" continues to be the song that Van Winkle is best known for internationally, although Van Winkle states that his American fans like his newer music better. According to Rolling Stone, the "Ice Ice Baby"–"Under Pressure" controversy is a landmark music copyright case, since it "sparked discussion about 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>
punitive actions taken in plagiarism cases". The magazine's Jordan Runtagh added: "Though [Vanilla Ice] paid the price, some argue that isn't enough to make up for the potential credibility lost by Queen and David Bowie, who are now linked to him through a collaboration they had no choice in joining." A live version of the song appeared on the album Extremely Live. "Ice Ice Baby" was rerecorded in a nu metal version titled "Too Cold". Originally intended to be released as a hidden track or B-side, "Too Cold" was featured on Van Winkle's 1998 album Hard to Swallow, and became a radio hit in some markets. In 2000, a remix titled "Ice Ice Baby 2001" was released in Europe as a single, with a newly produced music video. The remix generated new international interest in Van Winkle's music. VH1 and Blender ranked "Ice Ice Baby" fifth on its list of the "50 Most Awesomely Bad Songs Ever". It was also given the distinction by the Houston Press as being the worst song ever to emanate from Texas. In 1999, the song's music video was "retired" on the MTV special 25 Lame, in which Van Winkle himself appeared to destroy the video's master tape. Given a baseball bat, Van Winkle ended up destroying the show's set. However, in December 2007, VH1 ranked the song in 29th place of their 100 Greatest Songs of the 90's. In November 2011 MTV Dance ranked "Ice Ice Baby" at No. 71 in their list of "The 100 Biggest 90's Dance Anthems Of All Time". In 1991, Alvin and the Chipmunks released a cover version entitled "Ice Ice Alvin" for their album The Chipmunks Rock the House. "Weird Al" Yankovic included the chorus as the final song in "Polka Your Eyes Out", the polka medley from his 1992 album Off the Deep End. In 2004, the song was featured in the 13 Going on 30 in the movie. In 2010, the song was featured in the Glee episode "Bad Reputation" as performed by Will Schuester (Matthew Morrison). Track listings 1990 release 7" single "Ice Ice Baby" (radio mix) – 4:29 "Ice Ice Baby" (radio mix edit) – 3:49 12" maxi – U.S. "Ice Ice Baby" (radio mix) – 4:28 "Ice Ice Baby" (Miami drop mix) – 4:59 "Play That Funky Music" (radio mix) – 4:39 "Play That Funky Music" (instrumental mix) – 4:36 "Play That Funky Music" (a cappella mix) – 4:32 12" maxi / CD maxi – U.S. "Ice Ice Baby" (radio mix) – 4:28 "Ice Ice Baby" (Miami drop mix) – 4:59 "Ice Ice Baby" (Miami drop instrumental) – 4:59 "Ice Ice Baby" (a cappella mix) – 3:46 "Play That Funky Music" (radio mix) – 4:39 "Play That Funky Music" (instrumental mix) – 4:36 "Play That Funky Music" (a cappella mix) – 4:32 CD maxi – Europe "Ice Ice Baby" (radio edit) – 3:46 "Ice Ice Baby" (Miami drop mix) – 5:00 "Play That Funky Music" (radio mix) – 4:41 12" maxi – Europe "Ice Ice Baby" (club mix) – 5:02 "Ice Ice Baby" (radio mix) – 4:30 "Ice Ice Baby" (radio mix edit) – 3:49 12" maxi – UK "Ice Ice Baby" (Miami drop mix) – 4:58 "Ice Ice Baby" (instrumental mix) – 4:59 "It's a Party" – 4:39 "Ice Ice Baby" (radio mix) – 4:28 Cassette "Ice Ice Baby" (radio edit) – 3:46 "It's a Party" – 4:39 "Ice Ice Baby" (radio edit) – 3:46 "It's a Party" – 4:39 German CD Maxi "Ice Ice Baby" (Miami Drop Mix) "Ice Ice Baby" (Acapella Mix) "Ice Ice Baby" (Miami Drop Mix Instrumental) "Play
</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>
That Funky Music" (Acapella Mix) 2001 remixes 12" maxi "Ice Ice Baby 2001" (Gigi D'Agostino remix) – 7:17 "Ice Ice Baby 2001" (Funky 9ers club dub) – 4:53 "Ice Ice Baby 2001" (House of Wax club-mix) – 6:06 "Ice Ice Baby 2001" (Debart Style re-e-mix) – 6:42 CD maxi "Ice Ice Baby 2001" (House of Wax radio-mix) – 3:36 "Ice Ice Baby 2001" (Gigi D'Agostino remix-edit) – 3:45 "Ice Ice Baby 2001" (Silverwater & Shaw remix) – 3:42 "Ice Ice Baby 2001" (Prepay remix) – 3:54 "Ice Ice Baby 2001" (Steve Baltes remix) – 3:53 "Everytime (album version) (feat. 4BY4) – 3:58 2008 remixes 12" maxi "Ice Ice Baby 2008" (Mondo Electro remix) "Ice Ice Baby 2008" (7th Heaven House remix) "Ice Ice Baby 2008" (Rico NL Jumpstyle remix) "Ice Ice Baby 2008" (Mendezz and Andrew remix) Charts and sales Weekly charts Year-end charts Certifications |- |Australia | style="text-align:center;"|Platinum | style="text-align:center;"|70,000 |- |Austria | style="text-align:center;"|Gold | style="text-align:center;"|25,000 |- |Canada | style="text-align:center;"|Gold | style="text-align:center;"|50,000 |- |Germany | style="text-align:center;"|Gold | style="text-align:center;"|250,000 |- |- |Sweden | style="text-align:center;"|Gold | style="text-align:center;"|25,000 |- |United Kingdom | style="text-align:center;"|Platinum | style="text-align:center;"|600,000 |- |United States | style="text-align:center;"|Platinum | style="text-align:center;"|1,000,000 |} See also Under Pressure (Ice Ice Baby) References External links Category:1989 songs Category:1990 singles Category:Billboard Hot 100 number-one singles Category:Number-one singles in Australia Category:Number-one dance singles in Canada Category:European Hot 100 Singles number-one singles Category:Number-one singles in New Zealand Category:Irish Singles Chart number-one singles Category:UK Singles Chart number-one singles Category:Vanilla Ice songs Category:Songs involved in plagiarism controversies Category:Sampling controversies Category:Songs written by Vanilla Ice Category:Songs written by Freddie Mercury Category:Songs written by Brian May Category:Songs written by John Deacon Category:Songs written by Roger Taylor (Queen drummer) Category:Songs written by David Bowie Category:Songs involved in royalties controversies Category:SBK Records singles
</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>
Pedicab (band) Pedicab is a Filipino electronic rock group based in Manila formed in 2004. History Pedicab is composed of Diego Mapa ("Daddy Maps") on lead vocals and synths, Raimund Marasigan ("Sugar Raims") on synth bass and percussion, Mike Dizon ("Masterbeat") on drums and percussion, Jason Caballa ("J.Sonic") on guitar and cowbell, and RA Rivera ("Just Toni") on synths, sampler, and percussion. The band formed in August 2004 and played their first gig in September that year. In 2005, the band released their self-produced debut album entitled Tugish Takish under Vicor Music. Recorded in four days at Sound Creation Studios, Tugish Takish contains ten tracks, including the singles "Dizzy Boy," "Dito Tayo Sa Dilim," "Konti Na Lang," and "Bleached Streaks," with videos directed by RA Rivera. The band capped off the year by earning nominations for Album of the Year and Best New Artist at the NU 107 Rock Awards, winning in the latter category. A home demo entitled "Pa-Experience" was written for the compilation Rock Against The Round: Musicians Against The WTO in 2005. (A full band recording of the song was retitled "Pa-Taste" and appeared on the band's second album.) In December of that year, Pedicab performed in a street concert in Hong Kong (along with Noel Cabangon and the Radioactive Sago Project) organized by various Filipino NGOs in protest against the World Trade Organization. In July 2006, when the WTO reconvened in Geneva, Switzerland, the band flew there to perform as well. In 2007, Pedicab recorded a version of the VST and Company classic "Awitin Mo, Isasayaw Ko," which was included in the "Manila Sound" tribute compilation Hopia Mani Popcorn 2: The Best Of Manila Sound, released under Viva Records. The group signed a recording deal with MCA Music for their second album, Shinji Ilabas Mo Na Ang Helicopter, released in 2008. Produced by the band with assistance from Buddy Zabala and Mong Alcaraz, the album is a progression from the raw, dance-punk sound of their first record, with more varied beats and grooves, multiple guitar tracks, samples, and melodic (i.e. sung) vocals. The video for the carrier single "Ang Pusa Mo" was again directed by RA Rivera, and featured the band members being physically abused by beautiful women, including cosplay sisters Alodia Gosiengfiao and Ashley Gosiengfiao, current TV5 news anchor Shawn Yao, fashion blogger Tricia Gosingtian, and others. It won the Favorite Rock Video award at the 2009 Myx Music Awards. In March 2009, the band performed in Singapore for the annual Mosaic Music Festival, sharing the stage with Thai indie-pop quintet 15th Scenery and Chinese new wavers New Pants. In 2012, Pedicab released Kaya Mo Mag-Sando?, a six-track EP with the singles "Otomatik" and "Insekto." The EP showcases an even more evolved sound from the previous record, with several songs featuring acoustic guitars, various percussion instruments, and even more samples. That same year, the band contributed a new song, "Ubusan ng Lahi," recorded live for the soundtrack of the Marie Jamora-directed feature-length film and 2012 Cinemalaya Festival entry, Ang Nawawala. The band also appeared in the movie as themselves, performing the song live. The Ang Nawawalang Soundtrack was released on vinyl in 2013. In 2013, Pedicab wrote and recorded "Ang Dakilang Duwag Ng Katipunan" for the Rock Ed Philippines and Ballet Philippines production, Rock Supremo, a ballet performance about the life of Philippine revolutionary hero Andres Bonifacio, set to original music contributed by Filipino rock acts, including Ebe Dancel, Sandwich, Peryodiko, Peso Movement, Radioactive Sago Project, and others. The show ran for eight dates in September 2013. In 2014, Pedicab performed at the P-Fest UK
</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 Leeds and London, United Kingdom together with Top Junk, Kate Torralba, Squid 9, The Diegos, Sandwich (band) and Yano, as well as UK-based Filipino bands. The band spent most of 2015 writing new songs and are getting ready to go into the studio to record a new album in 2016. Members Diego Mapa (lead vocals/guitar/synths) Jason Caballa (lead guitar/backing vocals) Raimund Marasigan (synth bass/keyboards/keytar/backing vocals) Mike Dizon (drums/percussion) RA Rivera (sampler/programming/mastering/keyboard synths/percussion/backing vocals) Discography Studio albums Tugish Takish (2005; Vicor Records) Shinji Ilabas Mo Na Ang Helicopter (2008; MCA Music, Inc.) Kaya Mo Mag-Sando? (2012; MCA Music, Inc.) Remuda Triangle (2017; Soupstar Music) Compilations "Pa-Experience," Rock Against The Round: Musicians Against The WTO (2005; independent) "Awitin Mo, Isasayaw Ko" (VST and Company cover), Hopia Mani Popcorn 2: The Best Of Manila Sound (2007; Viva Records) "Sakto Sa Pasko," Sakto Sa Pasko (2008; Soupstar Music) "Ubusan Ng Lahi," Ang Nawawalang Soundtrack (2012; Indie Pop Records) "Ang Dakilang Duwag Ng Katipunan," Rock Supremo (2013; Rock Ed Philippines/Ballet Philippines) Singles "Dizzy Boy" "Dito Tayo sa Dilim" "Konti Na Lang" "Bleached Streaks" "Ang Pusa Mo" "FX" "Simulan Mo Na" "Otomatik" "Insekto" "What's The Algorithm?" References Category:Filipino rock music groups Category:Alternative dance musical groups Category:Electronic rock musical groups Category:Musical groups from Manila Category:MCA Music Inc. (Philippines) artists Category:Musical groups established in 2004
</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>
Raikoh Raikoh may refer to: Raikou (Pokémon) Minamoto no Yoritomo
</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>
RedLeg Rum RedLeg is a spiced rum first launched in the United Kingdom in 2012. It is owned by London-based spirits company Distil Plc, which also owns Blavod Black Vodka and Blackwoods Gin. It is distilled in the Caribbean, left to rest in old oak barrels and infused with Jamaican vanilla and ginger, then shipped to the UK for bottling. Its alcohol by volume is 37.5% (75 proof). The brand incorporates the motif of the Red Leg Hermit Crab that’s native to the Caribbean and is named for its bright legs. RedLeg is marketed globally as a premium spiced rum. External links RedLeg website Cocktail suggestion and review of RedLeg Spiced Rum The Fat Rum Pirate review of RedLeg References Category:Rums
</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>
Aduyevo Aduyevo () is the name of several rural localities in Russia: Aduyevo, Kaluga Oblast, a selo in Medynsky District of Kaluga Oblast Aduyevo, Moscow Oblast, a village in Yermolinskoye Rural Settlement of Istrinsky District in Moscow Oblast;
</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>
Mwewe River Mwewe River is a tributary river of the Thuli River in Zimbabwe. Category:Rivers of Zimbabwe
</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>
Novosphingobium endophyticum Novosphingobium endophyticum is a Gram-negative, rod-shaped and aerobic bacterium from the genus of Novosphingobium which has been isolated from the roots of the plant Glycyrrhiza uralensis from Yuli County in China. References External links Type strain of Novosphingobium endophyticum at BacDive - the Bacterial Diversity Metadatabase Category:Acidophiles Category:Bacteria described in 2016 Category:Sphingomonadales
</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>
Torpedo mackayana Torpedo mackayana, commonly known as the ringed torpedo, Western African torpedo, West African torpedo ray, McKay electric ray, or McKay's torpedo ray, is an electric ray species in the family Torpedinidae, which lives in shallow waters on the western coast of Africa. Characterized by rounded spiracles and white and brown spots, females grow to and males to . Taxonomy Torpedo mackayana was described in 1919 by Jan Metzelaar (1891–1929), a Dutch biologist. Description Torpedo mackayana has a round, "fleshy" disc, which has a slightly greater width than length. It has a gray-brown or rusty-brown upperside and a white underside. It is covered with small patches of brown or white. These patches may differ significantly in size and distribution throughout its body. It has a long, strong tail and two dorsal fins. The first dorsal fin is broad, while the second is slightly smaller and more slender. Its teeth are distributed in up to 38 rows and it has visible flaps on its nostrils. It has rounded spiracles, a feature that does not appear in any other species in the genus Torpedo. Torpedo mackayana is a small to medium-sized ray. Its growth and size was surveyed in the Coast of Senegal from 1994 to 1998, with the results having been published in 2001. This study showed that it has an average length of about and an average weight of at birth. Females typically reach sexual maturity at about and males reach it at . As an adult, females are larger than males. According to the survey, the length of adult females is and the length of adult males is . The largest female observed weighed and the largest male weighed . Ecology The reproductive cycle of Torpedo mackayana can take a year, and gestation can take nearly half a year. Ovulation happens in May or June, and young are born in August or September. Like other species in the genus Torpedo, the oocyte of the species is sometimes prevented from growing until birth. Females have two uteri and two ovaries. The eggs of the species, when fertilized, weigh about on average, and the oocytes weigh an average of . Torpedo mackayana has a coastal habitat and lives in depths of . It has been found in estuaries and on sea floors of mud or sand. Its prey consists of fish and smaller invertebrates. Little else is known about its biology. Distribution and conservation Torpedo mackayana lives in tropical waters of the eastern Atlantic Ocean, in 16 countries on the western coast of Africa. The most northern part of its range is Senegal, and the most southern part is Angola. It is likely caught by fisheries as a bycatch. Pollution and habitat destruction, due to coastal development, are also threats in part of its range. However, nothing is known about its population size or trend, except that it is not severely threatened. The adult population consists of more females than males, while young are more commonly male than female. No conservation actions are taking place for the species, and it is listed as data deficient on the IUCN Red List as of 2007. According to the IUCN, further research is required of the species' population before planning conservation actions. References Category:Torpedo (genus) Category:Fish described in 1919 Category:Fish of the East Atlantic Category:Marine fish of Africa
</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>
Aaron Walters Aaron Walters (August 7, 1809 – December 27, 1889) was a member of the Wisconsin State Assembly. Biography Walters was born on August 7, 1809 in Plattsburgh, New York. He married three times. First, to Orpha B. Griswold in 1831. They had three children before her death in 1839. Second, Walters married Rosanna Averill in 1840. They had two children before her death in 1858. In 1859, Walters married Mary Norman, a widowed mother of three. Through her mother, Mary was related to Enos T. Throop, a member of the United States House of Representatives and the 10th Governor of New York. He died of pneumonia on December 27, 1889. Career Walters was a member of the Assembly in 1857 and 1872. He was an unsuccessful candidate for the Assembly in 1859. Other positions Walters held include Chairman (similar to Mayor) of Eden, Fond du Lac County, Wisconsin and County Supervisor of Fond du Lac County, Wisconsin. He was a Democrat. References External links Category:People from Clinton County, New York Category:People from Eden, Wisconsin Category:Members of the Wisconsin State Assembly Category:Mayors of places in Wisconsin Category:County supervisors in Wisconsin Category:Wisconsin Democrats Category:1809 births Category:1889 deaths 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>
Matiota Kairo Matiota Kairo is a member of the Kiribati House of Assembly for the constituency of Tamana. References Category:Year of birth missing (living people) Category:Living people Category:Members of the House of Assembly of Kiribati Category:Pillars of Truth politicians Category:People from the Gilbert Islands
</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>
Dunne D.5 __NOTOC__ The Dunne D.5 was an experimental aircraft built in the United Kingdom in 1910. Designed by J. W. Dunne and built by Short Brothers at Leysdown, A tailless swept-wing biplane, the D.5 was the first aircraft built by his company, the Blair Atholl Aeroplane Syndicate Ltd. Like its military predecessors it was driven by twin pusher propellers, but it had a considerably more powerful engine. The D.5 first flew in the summer of 1910, Dunne having long ago literally dreamed of this flight. The D.5 was later certified as the first fixed-wing aircraft ever to achieve stability in flight, one of the official witnesses being Orville Wright. Development On leaving the Army Balloon Factory at Farnborough in 1909, J. W. Dunne set up a company, the Blair Atholl Aeroplane Syndicate Ltd., to continue developing his aircraft, none of which had yet flown under power. The D.5 was their first aircraft and it followed the basic design of the earlier D.1B and D.4 Army machines in having a tailless, swept biplane wing with pronounced wash-out. Like the others it was also driven by twin pusher propellers. However it differed in having a streamlined central nacelle or fuselage housing the pilot and engine, and the engine itself was a more powerful Green four-cylinder inline type. The controls were also unusual. Elevons at the wing tips provided all the control forces and were operated by two levers on either side of the cockpit. There was no rudder, with turning being coordinated by the aerodynamic design of the swept and washed-out wings. Endplate fins were fitted to the wings, to improve efficiency. Construction of the main airframe was contracted to Short Brothers. Operational history Following construction at Leysdown, the D.5 was taken to Eastchurch, the new site of the (now Royal) Aero Club and the Syndicate. Early trials were not encouraging, with the machine in its original form proving too heavy. The D.5 first flew, piloted by Dunne himself, in the summer of 1910. Dunne taxied to the top of a rise in the ground which lay downwind, turned the machine and took off downhill and into the wind. Dunne later recalled in his book An Experiment with Time that, as a child, he had experienced exactly this flight in a dream. The D.5 proved to be aerodynamically stable in flight. Two demonstration flights were made for the Royal Aero Club in December 1910, officially witnessed by the visiting Orville Wright and by Griffith Brewer. During the second flight, Dunne took his hands off the controls for an extended period, while he wrote a note on a piece of paper provided for him by Brewer. This note was the first ever documentary evidence of an aircraft's performance written in flight by the pilot himself. The D.5 was subsequently certified as the first fixed-wing aircraft ever to achieve stable flight. Specifications References Notes Bibliography Category:1910s British experimental aircraft Category:Dunne aircraft Category:Tailless aircraft Category:Biplanes Category:Single-engined twin-prop pusher aircraft Category:Aircraft with counter-rotating propellers Category:Aircraft first flown in 1910
</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>
Shelly Johnson (Twin Peaks) Shelly Johnson (, later known as Shelly Briggs), is a fictional character in the Twin Peaks franchise. She was created by the series creators Mark Frost and David Lynch and portrayed by Mädchen Amick. She is a main character in the original series, and has a supporting role in the prequel films Twin Peaks: Fire Walk with Me (1992) and Twin Peaks: The Missing Pieces (2014). Shelly returns in the 2017 revival series. In the original series, Shelly dropped out of high school to marry the trucker Leo Johnson, who began to abuse her after their marriage. Shelly works as a waitress at the Double R Diner and has an affair with Bobby Briggs. In Twin Peaks: The Return, it is revealed that Shelly married Bobby and still works at the Double R Diner. Her main storyline focuses on her trying to save their troubled daughter, Becky Burnett, from her own marriage to an abusive drug addict. Appearances In television Shelly is a waitress who dropped out of high school to marry the trucker Leo Johnson (Eric Da Re), who she had fallen for after he gave her attention. Once they got married, however, she realized Leo just "wanted a maid he didn't have to pay for", and that he became abusive whenever he didn't get his way. Shelly has been having an affair with Bobby Briggs (Dana Ashbrook), Laura's boyfriend. Leo finds out, and attempts to kill Shelly by tying her up inside the Packard Saw Mill and setting it on fire. However, Catherine Martell (Piper Laurie), another intended victim of the fire, frees Shelly and helps her escape. During the second season, Leo is in a catatonic state after being shot by Hank Jennings (Chris Mulkey). Shelly quits her job at the Double R Diner to care for Leo at home, where she and Bobby live together. When Leo awakens from his catatonia and tries to kill Shelly, Bobby comes and fights with Leo. Shelly stabs Leo in the leg, and Leo wanders off into the woods. Shelly later asks Norma for her old job at the Double R Diner, which Norma gives back to her. Shelly later receives fragments of poetry from former FBI agent Windom Earle (Kenneth Welsh). Shelly later enters the Miss Twin Peaks contest. Bobby later proposes to her, but she says no because she is still married to Leo. Leo is shot and murdered shortly after the events of the original series and autopsied by special agent Albert Rosenfield (Miguel Ferrer) on April 1, 1989; the cause had been five gunshots assumed by Rosenfield to have been fired by Earle, though tarantula bites had been present as well. Shelly and Bobby marry almost exactly one year later. Shelly returned in the 2017 Twin Peaks series with Mädchen Amick reprising her role. Shelly and Bobby divorced at some point prior to the revival, and they have an adult daughter, Becky (Amanda Seyfried). Shelly first appears in the second episode, drinking with friends at the Roadhouse, where she expresses concern for Becky, who is married to Steven Burnett (Caleb Landry Jones), an unemployed, abusive drug addict. She later has a nostalgic moment with James Hurley (James Marshall). In the fifth episode, Shelly is shown still working at the Double R Diner. Becky comes in and begs her for money, and Shelly reluctantly gives her all of the money that she has. Her boss Norma Jennings (Peggy Lipton), who has been described as a 'surrogate mother' figure for Shelly, has been observing the situation, comforts Shelly and tells her that she needs
</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 help her daughter while she still can. In the eleventh episode, while working at the Double R Diner, Shelly receives a phone call from a distressed Becky and runs out of the diner to go help her. When she arrives at Becky's trailer, her daughter snatches her keys and begins to drive off in her car. Shelly jumps onto the windshield until Becky flings her off. The owner of the trailer park, Carl Rodd (Harry Dean Stanton), gives her a ride to the Double R Diner as she calls Bobby to explain the situation. She and Bobby meet with Becky at the diner and make arrangements to protect her from Steven. As she hugs Becky she sees her new boyfriend Red (Balthazar Getty) - a drug dealer - outside the diner and kisses him. In the thirteenth episode, Shelly is working at the Double R Diner when Becky calls her expressing concern for Steven, as she hasn't seen him in two days. Shelly asks if she can call her later because she is working, and insists that Becky come to the diner so she can make her cherry pie. In film Shelly appears in the prequel films Twin Peaks: Fire Walk with Me (1992) and Twin Peaks: The Missing Pieces, the latter just consisting of unused footage from Fire Walk With Me. In Fire Walk With Me, Shelly is first seen working at the Double R Diner, when Norma asks her to assist Laura Palmer with the Meals on Wheels deliveries. When a distraught Laura tells Shelly that she cannot do the deliveries, Shelly is left to make the deliveries on her own. Later, Leo is shown verbally and physically abusing her, and forces her to wash the kitchen floor. When Bobby calls Leo, Shelly overhears the conversation and discovers that he owes him $5,000. In literature Shelly is referenced in the cookbook Damn Fine Cherry Pie: And Other Recipes from TV's Twin Peaks by Lindsey Bowden. Shelly is referenced in Jennifer Lynch's novel The Secret Diary of Laura Palmer. Told from the perspective of Laura Palmer, she wrote that she assumed that Shelly and Bobby were having an affair and discusses sleeping with Shelly's husband, Leo Johnson. Development Casting Mädchen Amick originally auditioned for the role of Donna Hayward, which ultimately went to Lara Flynn Boyle. However, the creators of the show were so impressed by her that they created the role of Shelly specifically for her. Reception In Pervert in the Pulpit: Morality in the Works of David Lynch, Jeff Johnson states that Shelly is one of the "salvageable characters" saying "Shelly, for instance, after so much abuse from Leo, gains a sense of self-confidence and a better self-image, giving up her scamming with Bobby and even entering the Miss Twin Peaks competition." Rolling Stone praised the distinctiveness of the character, stating that "Shelly Johnson was further away from the show's center than almost any of them. But that's kind of the point. Yes, she extricated herself from the high-school hell that consumed Laura by dropping out and getting married to brooding, drug-running trucker Leo Johnson. But that didn't save her from drudgery, boredom, betrayal, and abuse. It just moved the venue and raised the stakes." References External links Shelly Johnson Twin Peaks card Shelly Johnson at Twin Peaks (wiki) Category:Twin Peaks characters Category:Television characters introduced in 1990 Category:Teenage characters in film Category:Fictional waiting staff Category:Fictional domestic abuse victims
</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>
Appley Bridge meteorite The Appley Bridge meteorite is a meteorite that hit ground at Halliwell Farm in Appley Bridge, Lancashire, England at around 8:45 PM on Tuesday, 13 October 1914. After local residents saw a bolide, the meteorite was subsequently found in a farmer's field in the village the following day. It was below the surface of the field, with the appearance of burnt iron, and weighed almost . An article in Scientific News (No. 2588, 30 October 1914) stated "a small fragment which had been detached from the larger mass was put on view in a shop-window at Appley Bridge." A collection of letters, memoranda, and news-cuttings pertaining to the meteorite is held by the Natural History Museum Archives in London. In 2011, a fragment weighing less than an ounce and mounted in a one-inch plastic gem case was sold by auctioneers Lyon & Turnbull in Edinburgh. It was expected to sell for £250. See also Glossary of meteoritics Meteorite fall References Further reading Russell Parry (2014) The Appley Bridge Meteorite () Category:Borough of West Lancashire Category:Meteorite falls Category:1914 in England Category:Meteorites found in the United Kingdom
</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>
NBA 2K14 NBA 2K14 is a basketball video game developed by Visual Concepts and published by 2K Sports. The fifteenth installment in the series, it was released on October 1, 2013, for Microsoft Windows, PlayStation 3, and Xbox 360, and the PlayStation 4 and the Xbox One versions were released on the consoles' respective launch dates. LeBron James of the Miami Heat became the cover athlete, and also served as the music curator for the game. 2K Sports also released the DLC pack that as a bonus for pre-ordering the game, they would receive 10,000 VC, bonus content for "Path to Greatness" mode, adding James in the Blacktop mode and a signature skill for MyCareer mode. On August 14, 2013, 2K Sports announced that Crew Mode would return to this year's installment of NBA 2K. NBA 2K14 is the successor to NBA 2K13 and was succeeded by NBA 2K15 in the NBA 2K series. Development On July 2, 2013; 2K Sports and EuroLeague signed an exclusive multi-year deal to include teams in the game. This marks the first time EuroLeague teams have appeared in the NBA 2K series, similar to EA's NBA Live series, which included FIBA teams and also the U.S. national basketball team found in 2K13, beginning with Live 08 and ending with Live 10. LeBron James Soundtrack NBA 2K14 features soundtrack hand-picked by LeBron James. "2K is letting me take control of some important elements in NBA 2K14, including this year's soundtrack," James said in a statement. "I love that I'm the first NBA 2K cover athlete to choose the soundtrack." Big Krit - "Cool 2 be Southern" Cris Cab - "Paradise (On Earth)" Coldplay - "Lost!" Daft Punk (feat. Pharrell Williams) - "Get Lucky" Drake - "Started from the Bottom" Eminem - "Not Afraid" Fly Union - "Long Run" Gorillaz - "Clint Eastwood" Imagine Dragons - "Radioactive" Jadakiss (feat. Ayanna Irish) - "Can't Stop Me" Jay Z - "The Ruler's Back" John Legend (feat. Rick Ross) - "Who Do We Think We Are" Kanye West - "All of the Lights" Kendrick Lamar - "Now or Never" Macklemore & Ryan Lewis (feat. Ray Dalton) - "Can't Hold Us" Nas (feat. Puff Daddy) - "Hate Me Now" Phil Collins - "In The Air Tonight" Robin Thicke (feat. Pharrell) - "Blurred Lines" The Black Keys - "Elevator" The Black Keys - "Howlin For You" Reception NBA 2K 14 received "generally favorable" reviews from critics, according to review aggregator Metacritic. Ryan McCaffrey of IGN rated the game a 9.3 praising the gameplay saying, "With just enough new features, and tighter-than-ever gameplay, NBA 2K14 tops its predecessors and competitors... again". Tom McShea of GameSpot rated the game an 8.0 praising the game as well but criticizing designing of the game. USA Todays Brett Molina rated the game 3.5 out of 4 stars, noting that the game's new dribble and shoot mechanics may be simpler to newcomers, but frustrating to veteran gamers. Shortly after release, the game was the best-selling sports game on 8th-consoles. Additions and changes As well as the aforementioned new soundtrack and inclusion of EuroLeague teams (at the expense of the "Celeb" team, "Team USA", "Dream Team" and 2001–2002 Sacramento Kings classic squad), several new changes have been made since 2K13: "LeBron: Path to Greatness" mode, which is not found on the PlayStation 4 and Xbox One versions of the game, is reminiscent of a cut-down version of NBA 2K11s Jordan Challenge, where players get to control LeBron James' career as they attempt to win multiple rings (although his future is somewhat laid out by the game and choices
</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>
are limited). New signature skills, including an exclusive pre-order "LeBron Coast to Coast" skill (Speed, Quickness and Ball-Handling attributes are boosted 4 points and the shot penalty enforced for hitting a defender is decreased), One Man Fastbreak (a speed and handling increase on a fastbreak), Pick Dodger (player doesn't get caught on screens), Tenacious Rebounder (increase in rebounding skill), Flashy Passer (less likelihood of a "flashy pass" going astray - see below), Screen Outlet/Pick & Roll Maestro (skills which give boosts in pick and roll situations). a new way of passing by holding a trigger and pressing a control stick in the direction the player want to pass (includes blind passes). The "potential" stat is now limited and restricts the overall level a player can grow to, rather than increasing as the player improves (as happened in previous NBA 2K titles). This seems like a minor change but it has severely limited the amount by which players' stats can improve in Association Mode (and fixes one of the most often requested Association Mode flaws; too many players turning into superstars in Association Mode). More than 50 hours of new commentary from Kevin Harlan, Clark Kellogg and Steve Kerr, new draft night audio and new "MyCareer" voiceovers. A new "Training Camp" mode, which is there to guide players to master the new controls. Training Drills now have their own menu in My Career mode, meaning players don't have to return to the main menu (and back again) to do multiple training sessions. Always Online DRM added in patch requiring users to be connected to 2K servers to play MyCareer and MyGM modes. However the patch is only available on the PS4 and the Xbox One versions of the game and not on the PS3, Xbox 360 and PC versions of the game. References Category:2013 video games 14 Category:2K Sports games Category:Xbox One games Category:Xbox 360 games Category:PlayStation 4 games Category:PlayStation 3 games Category:Windows games Category:IOS games Category:Android (operating system) games Category:Video games developed in the United States Category:Video games set in 2013 Category:Video games set in 2014 Category:Video games set in Europe Category:Multiplayer and single-player video games Category:Take-Two Interactive games
</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>
Belica (river) The Belica (Serbian Cyrillic: Белица) is a left tributary of the Great Morava in Central Serbia. It is created by the confluence of smaller rivers Voljavica and Bešnjaja, flows through a deforested valley to the town Jagodina. Notes References External links Category:Rivers of Serbia
</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>
Nandipulam Nandipulam is a village in Varandarappilly panchayath in Thrissur district in the state of Kerala, India. Location Nandipulam is about Nine kilometer from Puthukad and Four kilometer from Kodakara Geography Temples Kumaranchira Bhagavathy Temple, Ramapuram Temple, Payyoorkkavu Temple, Mithrananthapuram Sree Krishna Temple, Edaleppilly Bhagavathy Temple, Chemmandaparambil Bhagavathy temple Etc.. Demographics India census, Nandipulam had a population of 5615 with 2689 males and 2926 females. References Category:Villages in Thrissur 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>
Morten Krogvold Morten Krogvold (born 3 May 1950) is a Norwegian photographer and writer. Krogvold is especially known for his portraits of artists, politicians and other celebrities. He has published numerous books, held numerous exhibitions. Early life and education Born in Oslo to calculator Gulbrand Leo Krogvold (1908–88) and mother Bodil Jacobsen (1913–88). Krogvold grew up in Lille Langerud in Oslo. At the age of twelve he acquired his own darkroom and became an enthusiastic amateur photographer. He received his professional training as a photographer at the Oslo Vocational School from 1973–1974, with Margaret Fosseide as his teacher. He was employed by the National Film Board from 1971 to 1981, a period that formed the basis for his development and interest in art. He captured images of interiors, sculpture, painting and furniture in museums and churches. Career At the end of the 1970s Morten Krogvold was active in the League of Free Photographers (FFF). He made his debut exhibition at the first photographic Spring exhibition 1976, he was secretary of the board for a short time, later juror and member of the committee. He attended the exhibition that FFF was presenting its members as guests at Autumn exhibition, and the exhibition Photography Here & Now at the Henie Onstad Art Center, both 1979. The same year he received the Norwegian Government's travel and study grants. The following year Krogvold as other photographer in history believed in the Autumn Exhibition, when he debuted with the solo exhibition in Norwegian Museum of Photography in Horten. During the 1980s, he broke with the Association and work increasingly with various assignments in advertising and documentaries. The missions have increasingly taken over and become an integral part of his exhibitions. Several of these are also presented in the book, such as Oslo Pictures in collaboration with Tove Nilsen (1988), or the White House (2000) for NRK. Morten Krogvold has since the 1980s had a major impact as a teacher in Norway and abroad. Especially at the Academy of Photography at Vågå he has with his ability to enthuse influenced and inspired numerous photography students. Krogvold was appointed Knight of the Order of St. Olav in 2005 and received Lom Municipality Culture for 2004. Bibliography 1982: Odd Hilt – Bilder gjennom 50 år (sm.m. H. Koefoed) 1983: Portretter/Portrait 1986: Det gjelder livet (for Norsk Kreftforening), Oslo 1986: Vi som er igjen 1988: Oslobilder (sm.m. T. Nilsen) 1989: Farmor dør (sm.m. P. Leer-Salvesen) 1993: Det avgjørende øyeblikket 1995: Du smiler til meg fra et falmet bilde (sm.m. M. Skjelbred) 1996: Bilder som speil (sm.m. K. M. Norderval) 1996: 100 portretter fra slutten av de første 100 år (for Norsk Filminstitutt) 1997: No Barriers (for Kon-Tiki Museet og Telenor) 1998: Sånn er eg (sm.m. I. H. Storrusten) 1998: Vi som blir igjen. Mennesker i møte med sorgen – og livet som går videre (sm.m. K. Ronge) 2000: Mann! Bibelske mannsbilder (sm.m. H. Rem) 2000: Det hvite hus (sm.m. S. Sandnes) 2000: Helse-Norge 2000 (for Norsk sykepleierforbund) 2001: Images References External links Category:1950 births Category:Living people Category:Norwegian photographers
</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>
Bralin, West Pomeranian Voivodeship Bralin () is a village in the administrative district of Gmina Kalisz Pomorski, within Drawsko County, West Pomeranian Voivodeship, in north-western Poland. It lies approximately east of Kalisz Pomorski, south-east of Drawsko Pomorskie, and east of the regional capital Szczecin. Before 1945 the area was part of Germany. For the history of the region, see History of Pomerania. The village has a population of 180. References Bralin
</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>
Albacete FS Albacete Fútbol Sala is a futsal club based in Albacete, the capital city of the province of Albacete in the autonomous community of Castile-La Mancha. The club was founded in 1984 and its pavilion is Pabellón Universitario with capacity of 1,200 seaters. The club's bat on the badge is similar to that of the city's football (soccer) team, Albacete Balompié. History Rise to Honor Division Descent to National A division Season to season 1 season in Primera División 14 seasons in Segunda División 3 season in Segunda División B 2 season in Tercera División External links Albacete FS Official Website Albacete Twitter Category:Futsal clubs in Spain Category:Sport in Albacete Category:Sports clubs established in 1984 Category:1984 establishments in Spain Category:Sports teams in Castilla–La Mancha
</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>
Georgia State Route 254 State Route 254 (SR 254) is a southwest-northeast state highway located in the northeastern part of the U.S. state of Georgia. It travels through portions of Hall and White counties. Route description SR 254 begins at an intersection with SR 284 (Main Street to the south, Shoal Creek Road to the north) in Clermont, in the northeastern part of Hall County. The route hads northeast through the northeastern part of Clermont along a northern section of Main Street. It makes a slight northwest jog before resuming its northeasterly routing. Just after resuming its trek to the northeast, it intersects US 129/SR 11 (Cleveland Highway). This intersection is just prior to entering the south-central part of White County. Farther to the northeast, the route passes Skitt Mountain Golf Course and Mossy Creek Campground. It then crosses over Dean, Mossy, White, and Flat Creeks before it intersects SR 384 just southeast of the unincorporated community of Leaf. Just after SR 384, the highway meets its eastern terminus, an intersection with SR 115, which is east of Leaf. No section of SR 254 is part of the National Highway System, a system of routes determined to be the most important for the nation's economy, mobility and defense. History SR 254 was established in 1949 along an alignment from US 129/SR 11 to its eastern terminus. In 1953, this entire section was paved. By 1963, the road was extended, and paved, along the rest of its current alignment, from Clermont to the intersection with US 129/SR 11. Major intersections See also References External links 254 Category:Transportation in Hall County, Georgia Category:Transportation in White County, Georgia
</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>
Euryops walterorum Euryops walterorum is a species of flowering plant in the family Asteraceae. It is found only in Namibia. References Category:Euryops Category:Flora of Namibia Category:Least concern plants Category:Taxonomy articles created by Polbot
</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>
Sinentomon Sinentomon is the only genus in the family Sinentomidae, in the hexapod order Protura. It contains three species found in China, Japan, and North Korea. Species Sinentomon chui Tuxen & Paik, 1982 Sinentomon erythranum Yin, 1965 Sinentomon yoroi Imadaté, 1977 References Category:Protura
</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>
Down in the Valley (film) Down in the Valley is a 2005 neo-western film starring Edward Norton, Evan Rachel Wood, David Morse and Rory Culkin. The film made its debut in the Un Certain Regard section at the 2005 Cannes Film Festival on May 13, and made its limited theatrical release in North America in May 5, 2006. Plot In the San Fernando Valley, rebellious teenager October "Tobe" takes a walk with her younger brother, Lonnie. The next day, Tobe goes to the beach with friends, and when they stop for gasoline, they are assisted by Harlan, a young man who affects a folksy, cowboy style. Tobe invites much-older Harlan to the beach. He accepts, which results in his losing his job. At the beach, they share a passionate kiss and, after returning to Harlan's house, they have sex. He takes her on a date, and the trio get something to eat. Later that night, they go on their "real" date, dancing and meeting up with Tobe's friends for another party, where Harlan takes drugs under the influence of Tobe. She returns home the next day; as she has returned home long after she was expected, Wade, her father, becomes enraged, and she retreats to her room. When she refuses to talk, he pounds on the door and leaves visible damage. Tobe continues to see Harlan. Her father's rage increases, and he shatters her bedroom window. The romantically involved couple ride a horse that supposedly belongs to one of Harlan's friends named Charlie. Upon returning, Charlie claims he has never met Harlan and that the horse was stolen. The couple are held in police custody until Wade comes to pick up Tobe. She tells Harlan that they should no longer see each other. Harlan, however, is persistent and takes Lonnie shooting without Wade's permission. Wade, who is armed, orders Harlan to leave his children alone. Mentally unstable, Harlan is evicted from his apartment after shooting at his reflection in a mirror, imagining a Wild West style "shoot-out". After an awkward incident at a local synagogue, where he is abruptly ushered out, he breaks into what is presumably the house of his father or foster father, who is revealed to be a Hasidic Jew. He leaves the letter he has been narrating throughout the film after taking multiple Jewish memorabilia, and the contents of a box, in a closet, inscribed with his name. He breaks into Tobe's house and packs a bag so that they can run away. When Tobe comes home to find him, she is dumbfounded, happy to see him at first. As she slowly realizes he is deranged, she tells him she does not want to leave her family and that he should go. As they argue, Harlan shoots her in the stomach. When Tobe's father returns home to find Tobe alone on her bed, barely alive, he suspects Harlan, who has failed in an attempt at calling 9-1-1 and run away. Wade rushes Tobe to the hospital, where she is attached to a breathing machine and remains in a coma. Harlan, who is covered in Tobe's blood, then shoots himself in the side to conceal Tobe's blood and also make it look like it was Wade who had shot them. Harlan finds Lonnie and convinces him that it was really Wade who shot Tobe, and that Harlan was wounded while trying to stop him. Tobe regains consciousness at the hospital and Wade realizes that Harlan has taken Lonnie. At night while Harlan and Lonnie are by a fire, Wade, Charlie and a detective named Sheridan arrive. Harlan
</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>
shoots Charlie before riding off with Lonnie. They stumble upon a Western film set where shooting has just begun. Wade and Sheridan arrive with two more cops. During the shootout, Harlan guns down detective Sheridan and one of the cops. Harlan and Lonnie escape to a construction site, where Wade finds them and another shootout ensues. Wade shoots Harlan to death to the horror of Lonnie. Later, Wade drives Tobe and Lonnie to a place where Tobe and Harlan had a pleasant day. Tobe is holding a box that contains her former lover's ashes. Her brother asks her what they should say about him. She replies, "Don't say anything, just think it," and scatters the ashes. Cast Writing Writer David Jacobson was inspired to write this film by his childhood in the San Fernando Valley. He commented that there was never much to do except throw things onto the highway (which possibly inspired a deleted scene from the film titled Don't Look), have dirt clod fights, and spending many hot summer days at the local cinema with friends, watching the same films over and over. One favorite was Butch Cassidy and the Sundance Kid, which he watched seventeen times. Jacobson also has noted that he and his sister were mild backgrounds for Tobe and Lonnie. The script was written with loose scenes, and is considered by Jacobson himself to be some of his lighter work. Critical reception American movie critic Roger Ebert expressed mixed feelings about the film in his review for the Chicago Sun Times. On the one hand, he noted several qualities "that make me happy to have seen it", especially the nuanced acting performance of Edward Norton and the "peculiar loneliness" of his character Harlan. On the other hand, Ebert took issue with the film's ending which he found to be implausible and driven too much by the abstract idea behind the plot instead of the characters in it. As of January 2012, the film holds a 51% rating on Rotten Tomatoes indicating mixed critical reception. References External links Official website (not active as of April 7, 2014) Category:2005 films Category:English-language films Category:2000s drama films Category:Films set in the San Fernando Valley Category:Films directed by David Jacobson Category:American independent films Category:American films Category:Revisionist Western
</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 Powlesland Alfred Powlesland (7 August 1875 – 25 February 1941) was an English cricket player. He was a member of the gold medal winning Great Britain cricket team at the 1900 Summer Olympics, the only time cricket has featured in the Olympic games. In the only match against France, Powlesland scored 10 in Great Britain's first innings and 4 in the second. He took two wickets in the French first innings and three wickets for 15 runs in their second. Little is known of his cricket career outside the Olympics except that he played for Exeter in 1900. References Olympic final scorecard Profiles of the Great Britain Olympic team Alfred Powlesland's profile at Sports Reference.com Category:English cricketers Category:English Olympic medallists Category:Olympic cricketers of Great Britain Category:Cricketers at the 1900 Summer Olympics Category:Olympic gold medallists for Great Britain Category:1875 births Category:1941 deaths Category:Medalists at the 1900 Summer Olympics
</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>
Unitary Unitary may refer to: Unitary construction, in automotive design a common term for unibody (unitary body/chassis) construction Lethal Unitary Chemical Agents and Munitions (Unitary), as chemical weapons opposite of Binary Unitarianism, in Christian doctrine, the belief in a "unitary God" as opposed to the concept of the Trinity Unitarian Universalism (sometimes called "Unitarianism"), an interfaith church which draws its name from an interfaith concept of "unitary God" types of political regions Unitary state Unitary authority Unitary enterprise, a type of government-owned corporation in Russia and Belarus Unitary executive theory, a theory of American constitutional law Mathematics E-unitary inverse semigroups Unitary matrix Unitary operator Unitary transformation Unitary group Unitary divisor Unitary representation Unital (disambiguation) Unitarity (physics)
</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>
Moussa Dagnogo Moussa Dagnogo (born 30 January 1972 in Paris) is a former French footballer who formerly worked as Head Coach of Carmel United Soccer Club. He is currently coaching premier and elite players at Jacksonville FC Playing career Dagnogo signed for Scottish club St Mirren as a trialist in February 2001, and scored the winning goal on his debut against Dundee. He was subsequently signed on a permanent contract. Dagnogo had a trial with Scottish Football League side, Clyde in December 2002. Coaching career In 2009, Dagnogo became the coach of Carmel United Soccer Club in Carmel, IN. At Carmel he had great success, leading his teams to strong runs at the state championship. From 2004 - 2011, he coached youth players in Indiana most recently at Carmel United Soccer Club. In 2010, Moussa’s U15 Girls won the State Cup. While at FC Pride, his U12 & U15 Girls were State Cup Semi-finalists. Moussa has also been Assistant Coach for Men’s & Women’s teams at St Joseph University. Moussa holds the National C Coaching License. In May 2011, Dagnogo relocated to Jacksonville, Florida and began coaching at the long established Jacksonville Youth Soccer Club. He assisted in training various teams and players, and was head coach of the Under 16 Boys FURY team. Under Dagnogo's leadership the team steadily climbed in state rankings and performed well in several tournaments in the Southeastern US. Dagnogo coached at a local high school in Jacksonville, FL from 2011 to 2013. During his tenure as head coach at Christ's Church Academy, he led his team to the divisional championships 3 straight years, and to the regional championships in his final season. Dagnogo then left to coach at the Jacksonville Football Club. In the 2015-16 season he was head coach of the Under 14 Boys Armada, leading them to the Quarterfinals in State cupand the Final of Florida Cup. References Category:1972 births Category:Living people Category:French footballers Category:Sportspeople from Paris Category:Angoulême-Soyaux Charente players Category:Racing Club de France Football players Category:Knattspyrnufélag Reykjavíkur players Category:Red Star F.C. players Category:Bristol Rovers F.C. players Category:St Mirren F.C. players Category:Association football forwards Category:Scottish Premier League players Category:Indiana Blast players Category:Cincinnati Kings players Category:French people of Ivorian descent Category:Indianapolis Braves players Category:Expatriate footballers in Scotland Category:Expatriate footballers in Portugal Category:French expatriate footballers Category:USL A-League players Category:USL Second Division players Category:US Sénart-Moissy players Category:Expatriate footballers in Iceland Category:Ivorian expatriates in Portugal Category:USL League Two 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>
Gary Keller (disambiguation) Gary Keller is an American basketball player/athlete. Gary Keller may also refer to: Gary Keller (saxophonist), American jazz and classical saxophonist Gary W. Keller, realtor and author
</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>
Oldtown Covered Bridge Oldtown Covered Bridge, near Oldtown, Kentucky, was listed on the National Register of Historic Places in 1976. It is located east of Kentucky Route 1, south of Oldtown. It brings Frazer Branch Road 750 over the Little Sandy River. It is a two-span Burr truss bridge. References Category:Covered bridges in Kentucky Category:National Register of Historic Places in Greenup County, Kentucky
</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>
Doña Francisquita Doña Francisquita is a zarzuela in three acts composed by Amadeo Vives to a Spanish libretto by Federico Romero and Guillermo Fernández-Shaw and based on Lope de Vega's play La discreta enamorada (The Ingenious Lover). With its colourful score and comic story of multiple love triangles which ends happily for the young lovers, Francisquita and Fernando, Doña Francisquita is considered a classic of the zarzuela genre, and Vives' masterpiece. Performance history Doña Francisquita premiered at the Teatro Apolo in Madrid on 17 October 1923. It became very popular, receiving over 5000 performances throughout the Spanish-speaking world in the years immediately following its premiere. The work is still regularly performed in Spain and Latin America, but like most works in this genre, has only rarely been seen elsewhere. It was, however, performed in French translation at Monte Carlo, Brussels and Vichy in 1934 and received a major production at Washington National Opera in 1998. There have been several film adaptations of Doña Francisquita, the most recent of which was directed by Ladislao Vajda and shown at the Cannes Film Festival in 1953 with Mirtha Legrand and Armando Calvo in the leading roles. The Act 1 soprano aria "Canción del ruiseñor" ("The Song of the Nightingale") and the Act 2 tenor aria "Por el humo se sabe donde está el fuego" ("We know where the fire is by the smoke") have been individually recorded or sung in recitals by many opera singers from Spanish-speaking countries. Roles Synopsis Time: 19th century Place: Madrid during the Carnival season. Francisquita is in love with student and would-be poet, Fernando. He, however, is infatuated with the fiery actress Aurora la Beltrana, who in turn is the lover of Lorenzo Pérez. When Francisquita tries to flirt with Fernando, he ignores her. His friend Cardona takes him to task and suggests that he would be happier with her than Aurora. Francisquita is then wooed by the Don Mathias, Fernando's father, in the presence of Doña Francisca, Francisquita's mother. At first, Doña Francisca mistakenly believes that she is the object of Don Mathias' affections. Francisquita pretends to accept the elderly man's advances, hoping to make Fernando jealous. At the urging of Cardona, Fernando also makes advances to Francisquita hoping to make Aurora jealous. Fernando's passion for Francisquita starts to grow. He is horrified that she might marry his father and become Doña Francisquita. He tells Aurora that she no longer has any power over him, which at last piques her interest in him. But it is too late, he is now completely in love with Francisquita. All is resolved after a series of misunderstandings and machinations involving Cardona disguising himself as a woman; Aurora's lover, Lorenzo Perez, challenging Fernando to a duel; and rumours that Fernando plans to marry Francisquita's mother to the subsequent horror of Don Mathias. Don Mathias eventually realizes he is too old for Francisquita and gives his blessing to the young couple. In the finale all celebrate the forthcoming wedding of Francisquita and Fernando by singing "Canción de la Juventud" ("Song of Youth"), a hymn to the youthful spirit of Madrid. Recordings María Bayo (Francisquita), Alfredo Kraus (Fernando), Raquel Pierotti (Aurora); Orquesta Sinfónica de Tenerife; Antoni Ros-Marbà (conductor). Label: Naïve Classique. María Rosa del Campo (Francisquita), Jaime Aragall (Fernando), Norma Lerer (Aurora); Orquesta Filarmonía de España; Lamberto Gardelli (conductor). Label: Alhambra/Columbia MCE 846-7 / SCE 946-7. Ainhoa Arteta (Francisquita), Plácido Domingo (Fernando), Linda Mirabal (Aurora); Orquesta Sinfónica de Sevilla; Miguel Roa (conductor). Label: Sony Classical. References Notes Sources Brown, Keith, Sunday Afternoon at the Opera, WWUH Radio, July–August 2008, Sadie, Stanley and Bashford,
</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>
Christina (eds) The New Grove Dictionary of Opera, Macmillan, 1992. Zinger, Pablo, "An Operatic Armada Sweeps Ashore." The New York Times, August 11, 1996. External links Christopher Webber, Detailed synopsis and analysis of Doña Francisquita on Zarzuela.net Category:Operas by Amadeo Vives Category:Spanish-language operas Category:1923 operas Category:Zarzuelas Category:Operas based on plays Category:Operas Category:Operas set in Spain
</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 tallest structures in China This list of the tallest buildings and structures in China ranks structures in China that stand at least tall by height. The list includes buildings located in Macau but not those found in Hong Kong, which are featured in their own list. Current Under construction This table ranks structures under construction with planned height at least by its planned height. It does not include structures that already reach their full height. See also List of tallest buildings in China References External links http://skyscraperpage.com/diagrams/?31045853 China, People's Republic of Tallest structures
</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>
Stole Dimitrievski Stole Dimitrievski (; born 25 December 1993) is a Macedonian professional footballer who plays for Spanish club Rayo Vallecano as a goalkeeper. After starting off at Rabotnički in 2010, he spent most of his career in Spain. Dimitrievski earned 28 caps for Macedonia at youth level, and made his senior international debut in 2015. Club career Early career Born in Kumanovo, Dimitrievski graduated from local giants FK Rabotnički's youth setup, and made his debut for the club on 24 October 2010, starting in a 2–0 home win against FK Vardar. Dimitrievski was Rabotnički's first-choice during the 2011–12 campaign, also appearing with the side in the season's UEFA Europa League. On 23 August 2011 he signed a contract with Italian Serie A side Udinese, with the deal only being effective in December. Granada On 30 December 2011, Dimitrievski was loaned to Cádiz CF. However, he only appeared with the reserves in Tercera División and moved to another reserve team, Granada CF B in the 2012 summer. On 12 October 2013, he was the last of three of the team's players sent off in a 2–1 loss at Atlético Sanluqueño CF for the season's Segunda División B. On 23 August 2014, after profiting from Roberto's suspension and Oier Olazábal's injury, Dimitrievski made his first team – and La Liga – debut, starting in a 2–1 home win against Deportivo de La Coruña. However, his spell at the club was mainly associated with the B-side. Gimnàstic On 16 August 2016, Dimitrievski signed a two-year deal with Segunda División side Gimnàstic de Tarragona, mainly as a cover to injured Manolo Reina. He made his debut for the Catalans on 7 September in the second round of the Copa del Rey, a 1–0 extra-time win at home to CD Numancia. After Sebastián Saja's poor run of form, Dimitrievski became a regular starter for Nàstic in October 2016, but his run only lasted until December as Reina returned from injury. On 13 August 2017, after the latter's departure, he extended his contract until 2020. Rayo Vallecano On 31 August 2018, Dimitrievski was loaned to Rayo Vallecano in the top flight, for one year. Initially a backup to Alberto García, he became the club's first-choice in November. On 31 January 2019, Dimitrievski signed a permanent three-and-a-half-year contract with the Madrid outskirts club. International career After representing Macedonia at under-17, under-19 and under-21 levels, Dimitrievski made his full international debut on 12 November 2015, starting in a 4–1 friendly routing of Montenegro at the Philip II Arena. References External links Profile at Macedonian Football Category:1993 births Category:Living people Category:Sportspeople from Kumanovo Category:Macedonian footballers Category:Association football goalkeepers Category:FK Rabotnički players Category:La Liga players Category:Segunda División B players Category:Tercera División players Category:Club Recreativo Granada players Category:Cádiz CF B players Category:Granada CF footballers Category:Gimnàstic de Tarragona footballers Category:Rayo Vallecano players Category:North Macedonia youth international footballers Category:North Macedonia international footballers Category:Macedonian expatriate footballers Category:Expatriate footballers in Spain Category:Macedonian expatriate sportspeople in Spain
</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>
TableCurve 2D TableCurve 2D is a linear and non-linear Curve fitting software package for engineers and scientists that automates the curve fitting process and in a single processing step instantly fits and ranks 3,600+ built-in frequently encountered equations enabling users to easily find the ideal model to their 2D data within seconds. Once the user has selected the best fit equation, they can output high-quality function and test programming codes or generate comprehensive reports and publication quality graphs. TableCurve 2D was originally developed by Ron Brown of AISN Software. The first version of TableCurve 2D was released in 1989. The first version was a DOS product. The first Windows based product was introduced in the last quarter of 1992. It was distributed by Jandel Scientific Software in the late 1980s but by January 2004, Systat Software acquired the exclusive worldwide rights from SPSS, Inc. to distribute SigmaPlot and other Sigma Series products. Systat Software is now based in San Jose, California. TableCurve 2D saves time by taking the endless trial and error out of curve fitting and that can help solve complex science and engineering problems faster. Related links SYSTAT PeakFit TableCurve 3D External links Systat Webpage TableCurve 2D Support Webpage Category:Plotting software Category:Regression and curve fitting software
</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>
Msingi Msingi is an administrative ward in the Mkalama District of the Singida Region of Tanzania. According to the 2002 census, the ward has a total population of 6,439. According to the 2012 census, the population had increased to 7,289. References Category:Populated places in Singida Region Category:Wards of Tanzania
</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>
Achievement (heraldry) An achievement, armorial achievement or heraldic achievement (historical: hatchment) in heraldry is a full display or depiction of all the heraldic components to which the bearer of a coat of arms is entitled. An achievement comprises not only the arms themselves displayed on the Escutcheon, the central element, but also the following elements surrounding it: Crest placed atop a: Torse (or Cap of Maintenance as a special honour) Mantling Helm of appropriate variety; if holder of higher rank than a baronet, issuing from a: Coronet or Crown (not used by baronets), of appropriate variety. Supporters (if the bearer is entitled to them, generally in modern usage not baronets), which may stand on a Compartment) Motto, if possessed Order, if possessed Badge, if possessed Coat of arms Sometimes the term "coat of arms" is used to refer to the full achievement, but this usage is wrong in the strict sense of heraldic terminology, as a coat of arms refers to a garment with the escutcheon or armorial achievement embroidered on it. Hatchment The ancient term used in place of "achievement" was "hatchment", being a corruption (through such historic forms as atcheament, achement, hathement, etc.) of the French achèvement, from the verb achever, a contraction of à chef venir ("to come to a head"), ultimately from Latin ad caput venire, "to come to a head", thus to reach a conclusion, accomplish, achieve. The word "hatchment" in its historical usage is thus identical in meaning and origin to the English heraldic term "achievement". However, in recent years the word "hatchment" has come to be used almost exclusively to denote "funerary hatchment", whilst "achievement" is now used in place of "hatchment" in a non-funereal context. An example of the historic use of "hatchment" in a non-funerary context to denote what is now termed "achievement" is in the statute of the Order of the Garter laid down by King Henry VIII (1509–1547) concerning the regulation of Garter stall plates:It is agreed that every knyght within the yere of his stallation shall cause to be made a scauchon of his armes and hachementis in a plate of metall suche as shall please him and that it shall be surely sett upon the back of his stall. References External links Category:Heraldry
</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>
Austrosteenisia Austrosteenisia is a genus of flowering plants in the legume family, Fabaceae. It belongs to the subfamily Faboideae. Category:Millettieae
</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. |