triple1 listlengths 3 3 | triple2 listlengths 3 3 | triple3 listlengths 3 3 | triple1_labels listlengths 3 3 | triple2_labels listlengths 3 3 | triple3_labels listlengths 3 3 | description stringlengths 9 15 | keep_triple bool 1
class | query_prompt stringlengths 3.24k 3.29k | query stringlengths 96 149 | entity_a stringlengths 5 18 | entity_b stringlengths 5 24 | rel_b stringclasses 9
values | prompt stringlengths 3.54k 3.63k | index int64 920 930 | raw_responses stringlengths 943 3.66k |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
[
"Q10322682",
"P40",
"Q892"
] | [
"Q76335458",
"P40",
"Q892"
] | [
"Q76335458",
"P25",
"Q76335461"
] | [
"Mabel Suffield",
"child",
"J. R. R. Tolkien"
] | [
"Arthur Reuel Tolkien",
"child",
"J. R. R. Tolkien"
] | [
"Arthur Reuel Tolkien",
"mother",
"Mary Jane Stow"
] | Q892__P40 | false | Task: Generate a Natural Language Query.
You will be given input in the following format:
Concept A: entity1
Concept B: entity2
Concept C: (concept B, rel, entity3)
Concept A and Concept C are connected via an independent concept B. Concept C is a property of Concept B, which the person is supposed to guess.
You... | What are different ways of connecting Mabel Suffield, and someone who is the mother of Arthur Reuel Tolkien? | Mabel Suffield | Mary Jane Stow | mother | Query: What are different ways of connecting Mabel Suffield, and someone who is the mother of Arthur Reuel Tolkien?
Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual triples (head, relationship, tail) forming a continuous chain that ... | 920 | {"final_response": "<answer>\n{\n \"1\": [\n [\"Mabel Suffield\", \"spouse\", \"Arthur Reuel Tolkien\"],\n [\"Arthur Reuel Tolkien\", \"mother\", \"Mary Jane Stow\"]\n ],\n \"2\": [\n [\"Mabel Suffield\", \"mother\", \"J.R.R. Tolkien\"],\n [\"J.R.R. Tolkien\", \"father\", \"Arthur Reuel Tolkien\"],\n ... |
[
"Q2424867",
"P162",
"Q10338974"
] | [
"Q3376853",
"P162",
"Q10338974"
] | [
"Q10338974",
"P166",
"Q708731"
] | [
"Thomas Häberle",
"producer",
"Suely in the Sky"
] | [
"Peter Rommel",
"producer",
"Suely in the Sky"
] | [
"Suely in the Sky",
"award received",
"German Film Award"
] | Q10338974__P162 | false | Task: Generate a Natural Language Query.
You will be given input in the following format:
Concept A: entity1
Concept B: entity2
Concept C: (concept B, rel, entity3)
Concept A and Concept C are connected via an independent concept B. Concept C is a property of Concept B, which the person is supposed to guess.
You... | What are different ways of connecting Suely in the Sky, a notable figure in the film industry, and someone who has received the German Film Award? | Suely in the Sky | German Film Award | award received | Query: What are different ways of connecting Suely in the Sky, a notable figure in the film industry, and someone who has received the German Film Award?
Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual triples (head, relationship, ... | 921 | {"final_response": "```json\n<answer>\n{\n \"1\": [\n [\"Suely in the Sky\", \"producer\", \"Thomas Kufus\"],\n [\"Thomas Kufus\", \"award received\", \"German Film Award\"]\n ],\n \"2\": [\n [\"Suely in the Sky\", \"director\", \"Karim A\u00efnouz\"],\n [\"Karim A\u00efnouz\", \"directed\", \"Firebrand\... |
[
"Q1987718",
"P463",
"Q20672"
] | [
"Q3812070",
"P463",
"Q20672"
] | [
"Q3812070",
"P136",
"Q316930"
] | [
"Peredur ap Gwynedd",
"member of",
"Pendulum"
] | [
"KJ Sawka",
"member of",
"Pendulum"
] | [
"KJ Sawka",
"genre",
"dance music"
] | Q20672__P463 | false | Task: Generate a Natural Language Query.
You will be given input in the following format:
Concept A: entity1
Concept B: entity2
Concept C: (concept B, rel, entity3)
Concept A and Concept C are connected via an independent concept B. Concept C is a property of Concept B, which the person is supposed to guess.
You... | What are different ways of connecting Peredur ap Gwynedd, a figure from Welsh mythology, and someone who is associated with the genre of dance music? | Peredur ap Gwynedd | dance music | genre | Query: What are different ways of connecting Peredur ap Gwynedd, a figure from Welsh mythology, and someone who is associated with the genre of dance music?
Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual triples (head, relationshi... | 922 | {"final_response": "<answer>\n{\n \"1\": [\n [\"Peredur ap Gwynedd\", \"member of\", \"Pendulum\"],\n [\"Pendulum\", \"genre\", \"dance music\"]\n ],\n \"2\": [\n [\"Peredur ap Gwynedd\", \"touring member of\", \"Faithless\"],\n [\"Faithless\", \"genre\", \"dance music\"]\n ],\n \"3\": [\n [\"Peredu... |
[
"Q18037046",
"P2293",
"Q62736"
] | [
"Q18055508",
"P2293",
"Q62736"
] | [
"Q18055508",
"P5572",
"Q17145"
] | [
"MCF2L",
"genetic association",
"osteoarthritis"
] | [
"NACA2",
"genetic association",
"osteoarthritis"
] | [
"NACA2",
"expressed in",
"sperm"
] | Q62736__P2293 | false | Task: Generate a Natural Language Query.
You will be given input in the following format:
Concept A: entity1
Concept B: entity2
Concept C: (concept B, rel, entity3)
Concept A and Concept C are connected via an independent concept B. Concept C is a property of Concept B, which the person is supposed to guess.
You... | What are different ways of connecting MCF2L, a gene, and a substance that is expressed in sperm? | MCF2L | sperm | expressed in | Query: What are different ways of connecting MCF2L, a gene, and a substance that is expressed in sperm?
Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual triples (head, relationship, tail) forming a continuous chain that begins with ... | 923 | {"final_response": "<answer>\n{\n \"1\": [\n [\"MCF2L\", \"interacts with\", \"RHOA\"],\n [\"RHOA\", \"expressed in\", \"sperm\"]\n ],\n \"2\": [\n [\"MCF2L\", \"interacts with\", \"CDC42\"],\n [\"CDC42\", \"expressed in\", \"sperm\"]\n ],\n \"3\": [\n [\"MCF2L\", \"interacts with\", \"RAC1\"],\n ... |
[
"Q692945",
"P463",
"Q11890"
] | [
"Q1228158",
"P463",
"Q11890"
] | [
"Q1228158",
"P106",
"Q177220"
] | [
"Arne Zank",
"member of",
"Tocotronic"
] | [
"Dirk von Lowtzow",
"member of",
"Tocotronic"
] | [
"Dirk von Lowtzow",
"occupation",
"singer"
] | Q11890__P463 | false | Task: Generate a Natural Language Query.
You will be given input in the following format:
Concept A: entity1
Concept B: entity2
Concept C: (concept B, rel, entity3)
Concept A and Concept C are connected via an independent concept B. Concept C is a property of Concept B, which the person is supposed to guess.
You... | What are different ways of connecting Arne Zank, a notable figure in the music scene, and someone who is a singer? | Arne Zank | singer | occupation | Query: What are different ways of connecting Arne Zank, a notable figure in the music scene, and someone who is a singer?
Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual triples (head, relationship, tail) forming a continuous chain... | 924 | {"final_response": "<answer>\n{\n \"1\": [\n [\"Arne Zank\", \"member of\", \"Tocotronic\"],\n [\"Tocotronic\", \"has member\", \"Dirk von Lowtzow\"],\n [\"Dirk von Lowtzow\", \"occupation\", \"singer\"]\n ],\n \"2\": [\n [\"Arne Zank\", \"instrument\", \"Drum kit\"],\n [\"Drum kit\", \"played by\", \... |
[
"Q14913379",
"P2293",
"Q128581"
] | [
"Q18034329",
"P2293",
"Q128581"
] | [
"Q18034329",
"P684",
"Q18297015"
] | [
"ABCC4",
"genetic association",
"breast cancer"
] | [
"SLC4A7",
"genetic association",
"breast cancer"
] | [
"SLC4A7",
"ortholog",
"Slc4a7"
] | Q128581__P2293 | false | Task: Generate a Natural Language Query.
You will be given input in the following format:
Concept A: entity1
Concept B: entity2
Concept C: (concept B, rel, entity3)
Concept A and Concept C are connected via an independent concept B. Concept C is a property of Concept B, which the person is supposed to guess.
You... | What are different ways of connecting ABCC4, a gene involved in transport processes, and a substance that is an ortholog of SLC4A7? | ABCC4 | Slc4a7 | ortholog | Query: What are different ways of connecting ABCC4, a gene involved in transport processes, and a substance that is an ortholog of SLC4A7?
Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual triples (head, relationship, tail) forming a... | 925 | {"final_response": "<answer>\n{\n \"1\": [\n [\"ABCC4\", \"expressed in\", \"Kidney\"],\n [\"Kidney\", \"expresses\", \"SLC4A7\"],\n [\"SLC4A7\", \"ortholog\", \"Slc4a7\"]\n ],\n \"2\": [\n [\"ABCC4\", \"found in taxon\", \"Homo sapiens\"],\n [\"Homo sapiens\", \"has gene\", \"SLC4A7\"],\n [\"SLC4A... |
[
"Q20817012",
"P2175",
"Q47790"
] | [
"Q5198674",
"P2175",
"Q47790"
] | [
"Q5198674",
"P2868",
"Q50430093"
] | [
"metocurine iodide",
"medical condition treated",
"tetanus"
] | [
"cyclobenzaprine",
"medical condition treated",
"tetanus"
] | [
"cyclobenzaprine",
"subject has role",
"central muscle relaxants"
] | Q47790__P2175 | false | Task: Generate a Natural Language Query.
You will be given input in the following format:
Concept A: entity1
Concept B: entity2
Concept C: (concept B, rel, entity3)
Concept A and Concept C are connected via an independent concept B. Concept C is a property of Concept B, which the person is supposed to guess.
You... | What are different ways of connecting metocurine iodide, a chemical compound, and a substance that has a role as a central muscle relaxant? | metocurine iodide | central muscle relaxants | subject has role | Query: What are different ways of connecting metocurine iodide, a chemical compound, and a substance that has a role as a central muscle relaxant?
Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual triples (head, relationship, tail) f... | 926 | {"final_response": "<answer>\n{\n \"1\": [\n [\n \"metocurine iodide\",\n \"interacts with\",\n \"diazepam\"\n ],\n [\n \"diazepam\",\n \"subject has role\",\n \"central muscle relaxants\"\n ]\n ],\n \"2\": [\n [\n \"metocurine iodide\",\n \"interacts with\",\n ... |
[
"Q18043155",
"P2293",
"Q11081"
] | [
"Q18045297",
"P2293",
"Q11081"
] | [
"Q18045297",
"P5572",
"Q975054"
] | [
"ARHGAP20",
"genetic association",
"Alzheimer's disease"
] | [
"EXOC4",
"genetic association",
"Alzheimer's disease"
] | [
"EXOC4",
"expressed in",
"tibialis anterior muscle"
] | Q11081__P2293 | false | Task: Generate a Natural Language Query.
You will be given input in the following format:
Concept A: entity1
Concept B: entity2
Concept C: (concept B, rel, entity3)
Concept A and Concept C are connected via an independent concept B. Concept C is a property of Concept B, which the person is supposed to guess.
You... | What are different ways of connecting ARHGAP20, a gene, and a substance that is expressed in the tibialis anterior muscle? | ARHGAP20 | tibialis anterior muscle | expressed in | Query: What are different ways of connecting ARHGAP20, a gene, and a substance that is expressed in the tibialis anterior muscle?
Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual triples (head, relationship, tail) forming a continuo... | 927 | {"final_response": "<answer>\n{\n \"1\": [\n [\"ARHGAP20\", \"expressed in\", \"tibialis anterior muscle\"]\n ],\n \"2\": [\n [\"ARHGAP20\", \"interacts with\", \"RHOA\"],\n [\"RHOA\", \"expressed in\", \"tibialis anterior muscle\"]\n ],\n \"3\": [\n [\"ARHGAP20\", \"interacts with\", \"RAC1\"],\n [... |
[
"Q552748",
"P463",
"Q32825"
] | [
"Q312450",
"P463",
"Q32825"
] | [
"Q312450",
"P740",
"Q160642"
] | [
"Antony Costa",
"member of",
"Blue"
] | [
"Duncan James",
"member of",
"Blue"
] | [
"Duncan James",
"location of formation",
"Salisbury"
] | Q32825__P463 | false | Task: Generate a Natural Language Query.
You will be given input in the following format:
Concept A: entity1
Concept B: entity2
Concept C: (concept B, rel, entity3)
Concept A and Concept C are connected via an independent concept B. Concept C is a property of Concept B, which the person is supposed to guess.
You... | What are different ways of connecting Antony Costa, the English singer, and someone who is associated with the formation in Salisbury? | Antony Costa | Salisbury | location of formation | Query: What are different ways of connecting Antony Costa, the English singer, and someone who is associated with the formation in Salisbury?
Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual triples (head, relationship, tail) formin... | 928 | {"final_response": "<answer>\n{\n \"1\": [\n [\"Antony Costa\", \"participated in\", \"I'm a Celebrity...Get Me Out of Here!\"],\n [\"Christopher Biggins\", \"participated in\", \"I'm a Celebrity...Get Me Out of Here!\"],\n [\"Christopher Biggins\", \"acted at\", \"Salisbury Playhouse\"],\n [\"Salisbury Pl... |
[
"Q3626966",
"P463",
"Q146027"
] | [
"Q3624971",
"P463",
"Q146027"
] | [
"Q3624971",
"P54",
"Q19589"
] | [
"Harry Styles",
"member of",
"One Direction"
] | [
"Louis Tomlinson",
"member of",
"One Direction"
] | [
"Louis Tomlinson",
"member of sports team",
"Doncaster Rovers F.C."
] | Q146027__P463 | false | Task: Generate a Natural Language Query.
You will be given input in the following format:
Concept A: entity1
Concept B: entity2
Concept C: (concept B, rel, entity3)
Concept A and Concept C are connected via an independent concept B. Concept C is a property of Concept B, which the person is supposed to guess.
You... | What are different ways of connecting Harry Styles, the British singer, and someone who is a member of the Doncaster Rovers F.C. sports team? | Harry Styles | Doncaster Rovers F.C. | member of sports team | Query: What are different ways of connecting Harry Styles, the British singer, and someone who is a member of the Doncaster Rovers F.C. sports team?
Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual triples (head, relationship, tail)... | 929 | {"final_response": "<answer>\n{\n \"1\": [\n [\n \"Harry Styles\",\n \"bandmate of\",\n \"Louis Tomlinson\"\n ],\n [\n \"Louis Tomlinson\",\n \"member of sports team\",\n \"Doncaster Rovers F.C.\"\n ]\n ],\n \"2\": [\n [\n \"Harry Styles\",\n \"member of\",\n ... |
[
"Q1392744",
"P161",
"Q52162432"
] | [
"Q1392744",
"P161",
"Q4931077"
] | [
"Q4931077",
"P106",
"Q2722764"
] | [
"The Wolf of Wall Street",
"cast member",
"Stephanie Kurtzuba"
] | [
"The Wolf of Wall Street",
"cast member",
"Bo Dietl"
] | [
"Bo Dietl",
"occupation",
"radio personality"
] | Q1392744__P161 | false | Task: Generate a Natural Language Query.
You will be given input in the following format:
Concept A: entity1
Concept B: entity2
Concept C: (concept B, rel, entity3)
Concept A and Concept C are connected via an independent concept B. Concept C is a property of Concept B, which the person is supposed to guess.
You... | What are different ways of connecting Stephanie Kurtzuba, an actress, and someone who is known for their work as a radio personality? | Stephanie Kurtzuba | radio personality | occupation | Query: What are different ways of connecting Stephanie Kurtzuba, an actress, and someone who is known for their work as a radio personality?
Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual triples (head, relationship, tail) forming... | 930 | {"final_response": "<answer>\n{\n \"1\": [\n [\"Stephanie Kurtzuba\", \"acted in\", \"The Wolf of Wall Street\"],\n [\"The Wolf of Wall Street\", \"features actor\", \"Jonah Hill\"],\n [\"Jonah Hill\", \"interviewed by\", \"Howard Stern\"],\n [\"Howard Stern\", \"occupation\", \"radio personality\"]\n ],\... |
Dataset: connections-dev/connection_queries_jan12
This dataset was generated using the inference script with the following configuration:
Inference Parameters
Model Configuration
- Model Name:
gemini-3.1-pro-preview - Server URL:
Not specified - API Key:
Not provided - Request Timeout:
120 seconds
Query Configuration
- Query Type:
natural - Query Column:
query - Sampling Type:
original
Generation Parameters
- Temperature:
0.7 - Max Tokens:
16384 - Number of Runs:
1 - Chunk Size:
Entire dataset - Effort Level:
medium
Feature Flags
- Step by Step Reasoning:
False - Reasoning Enabled:
False - No Curator:
False - Index Filter:
None (all samples) - Start Index:
920 - End Index:
935
Data Information
- Input File:
connections-dev/connection_queries_jan12 - Number of Samples:
11 - Output Filename:
connection_queries_jan12__natural__s920__e935__original__1__medium__0.7__16384__gemini-3_1-pro-preview.jsonl
Dataset Structure
The dataset contains the following key columns:
query: The original queryentity_a: Starting entityentity_b: Target entityrel_b: Target relationshippath_prediction: Generated connection pathsprompt_for_path: Prompts used for path generation
Usage
You can load this dataset using:
from datasets import load_dataset
dataset = load_dataset("s920_e935_original_1_medium_0.7_16384_gemini-3_1-pro-preview__raw")
Generation Details
This dataset was generated on 2026-03-12 13:11:32 using the inference pipeline with the above configuration.
- Downloads last month
- 14