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 8 15 | keep_triple bool 1
class | query_prompt stringlengths 3.24k 3.3k | query stringlengths 87 171 | entity_a stringlengths 5 22 | entity_b stringlengths 5 45 | rel_b stringlengths 5 28 | prompt stringlengths 3.55k 3.65k | raw_responses stringlengths 2.64k 13.8k |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
[
"Q3705835",
"P54",
"Q337758"
] | [
"Q7816222",
"P54",
"Q337758"
] | [
"Q7816222",
"P19",
"Q65"
] | [
"Derek Carrier",
"member of sports team",
"San Francisco 49ers"
] | [
"Tom Holmoe",
"member of sports team",
"San Francisco 49ers"
] | [
"Tom Holmoe",
"place of birth",
"Los Angeles"
] | Q337758__P54 | 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 Derek Carrier, the professional football player, and someone who was born in Los Angeles? | Derek Carrier | Los Angeles | place of birth | Query: What are different ways of connecting Derek Carrier, the professional football player, and someone who was born in Los Angeles?
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 con... | {"final_response": "<think>\n\n</think>\n\n<answer>\n{\n \"1\": [\n [\"Derek Carrier\", \"member of sports team\", \"Las Vegas Raiders\"],\n [\"Las Vegas Raiders\", \"player roster includes\", \"DeSean Jackson\"],\n [\"DeSean Jackson\", \"place of birth\", \"Los Angeles\"]\n ],\n \"2\": [\n [\"Derek Carr... |
[
"Q649593",
"P40",
"Q76"
] | [
"Q766106",
"P40",
"Q76"
] | [
"Q766106",
"P69",
"Q6969036"
] | [
"Barack Obama Sr.",
"child",
"Barack Obama"
] | [
"Stanley Ann Dunham",
"child",
"Barack Obama"
] | [
"Stanley Ann Dunham",
"educated at",
"Nathan Eckstein Middle School"
] | Q76__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 Barack Obama Sr., and someone who was educated at Nathan Eckstein Middle School? | Barack Obama Sr. | Nathan Eckstein Middle School | educated at | Query: What are different ways of connecting Barack Obama Sr., and someone who was educated at Nathan Eckstein Middle School?
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 c... | {"final_response": "<think>\n\n</think>\n\n<answer>\n{\n \"1\": [\n [\n \"Barack Obama Sr.\",\n \"father of\",\n \"Barack Obama\"\n ],\n [\n \"Barack Obama\",\n \"awarded\",\n \"Presidential Medal of Freedom\"\n ],\n [\n \"Presidential Medal of Freedom\",\n \"awar... |
[
"Q5199066",
"P2175",
"Q41861"
] | [
"Q388801",
"P2175",
"Q41861"
] | [
"Q388801",
"P769",
"Q241006"
] | [
"cyclothiazide",
"medical condition treated",
"arterial hypertension"
] | [
"furosemide",
"medical condition treated",
"arterial hypertension"
] | [
"furosemide",
"significant drug interaction",
"desmopressin"
] | Q41861__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 cyclothiazide, a medication, and a substance that has a significant drug interaction with desmopressin? | cyclothiazide | desmopressin | significant drug interaction | Query: What are different ways of connecting cyclothiazide, a medication, and a substance that has a significant drug interaction with desmopressin?
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)... | {"final_response": "<think>\n\n</think>\n\n<answer>\n{\n \"1\": [\n [\n \"cyclothiazide\",\n \"chemical subclass\",\n \"benzothiadiazine\"\n ],\n [\n \"benzothiadiazine\",\n \"includes\",\n \"chlorothiazide\"\n ],\n [\n \"chlorothiazide\",\n \"significant drug int... |
[
"Q173123",
"P54",
"Q169898"
] | [
"Q281294",
"P54",
"Q169898"
] | [
"Q281294",
"P2522",
"Q4623085"
] | [
"Jacky Ickx",
"member of sports team",
"Scuderia Ferrari"
] | [
"Jules Bianchi",
"member of sports team",
"Scuderia Ferrari"
] | [
"Jules Bianchi",
"victory",
"2011 Yas Marina Circuit GP2 Asia Series round"
] | Q169898__P54 | 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 Jacky Ickx, the Belgian racing driver, and someone who achieved victory at the 2011 Yas Marina Circuit GP2 Asia Series round? | Jacky Ickx | 2011 Yas Marina Circuit GP2 Asia Series round | victory | Query: What are different ways of connecting Jacky Ickx, the Belgian racing driver, and someone who achieved victory at the 2011 Yas Marina Circuit GP2 Asia Series round?
Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual triples (hea... | {"final_response": "<think>\n\n</think>\n\n<answer>\n{\n \"1\": [\n [\n \"Jacky Ickx\",\n \"father of\",\n \"Vanina Ickx\"\n ],\n [\n \"Vanina Ickx\",\n \"competed in\",\n \"24 Hours of Le Mans\"\n ],\n [\n \"24 Hours of Le Mans\",\n \"competitor\",\n \"Romai... |
[
"Q18040268",
"P2293",
"Q131755"
] | [
"Q18047740",
"P2293",
"Q131755"
] | [
"Q18047740",
"P5572",
"Q975459"
] | [
"COMMD10",
"genetic association",
"bipolar disorder"
] | [
"MYO18B",
"genetic association",
"bipolar disorder"
] | [
"MYO18B",
"expressed in",
"vastus lateralis muscle"
] | Q131755__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 COMMD10, a protein involved in cellular processes, and a substance that is expressed in the vastus lateralis muscle? | COMMD10 | vastus lateralis muscle | expressed in | Query: What are different ways of connecting COMMD10, a protein involved in cellular processes, and a substance that is expressed in the vastus lateralis 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, relati... | {"final_response": "<think>\n\n</think>\n\n<answer>\n{\n \"1\": [\n [\n \"COMMD10\",\n \"physically interacts with\",\n \"CUL2\"\n ],\n [\n \"CUL2\",\n \"expressed in\",\n \"vastus lateralis muscle\"\n ]\n ],\n \"2\": [\n [\n \"COMMD10\",\n \"physically interact... |
[
"Q18052870",
"P2293",
"Q11081"
] | [
"Q17825605",
"P2293",
"Q11081"
] | [
"Q17825605",
"P5572",
"Q2478147"
] | [
"GLIS3",
"genetic association",
"Alzheimer's disease"
] | [
"BIN1",
"genetic association",
"Alzheimer's disease"
] | [
"BIN1",
"expressed in",
"glutes"
] | 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 GLIS3, a gene associated with various biological functions, and a substance that is expressed in the glutes? | GLIS3 | glutes | expressed in | Query: What are different ways of connecting GLIS3, a gene associated with various biological functions, and a substance that is expressed in the glutes?
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, ... | {"final_response": "<think>\n\n</think>\n\n<answer>\n{\n \"1\": [\n [\n \"GLIS3\",\n \"associated with\",\n \"Type 1 diabetes\"\n ],\n [\n \"Type 1 diabetes\",\n \"involves dysfunction of\",\n \"Insulin\"\n ],\n [\n \"Insulin\",\n \"regulates glucose uptake in\",\... |
[
"Q1025096",
"P162",
"Q970867"
] | [
"Q1025096",
"P162",
"Q6988354"
] | [
"Q6988354",
"P69",
"Q168000"
] | [
"Caddyshack II",
"producer",
"Peter Guber"
] | [
"Caddyshack II",
"producer",
"Neil Canton"
] | [
"Neil Canton",
"educated at",
"American University"
] | Q1025096__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 Peter Guber, the American film producer, and someone who was educated at American University? | Peter Guber | American University | educated at | Query: What are different ways of connecting Peter Guber, the American film producer, and someone who was educated at American University?
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... | {"final_response": "<think>\n\n</think>\n\n<answer>\n{\n \"1\": [\n [\n \"Peter Guber\",\n \"co-owner of\",\n \"Golden State Warriors\"\n ],\n [\n \"Golden State Warriors\",\n \"head coach\",\n \"Steve Kerr\"\n ],\n [\n \"Steve Kerr\",\n \"child of\",\n \"Mal... |
[
"Q184577",
"P166",
"Q38104"
] | [
"Q9021",
"P166",
"Q38104"
] | [
"Q9021",
"P802",
"Q60347"
] | [
"Walter Houser Brattain",
"award received",
"Nobel Prize in Physics"
] | [
"Max Planck",
"award received",
"Nobel Prize in Physics"
] | [
"Max Planck",
"student",
"Walter H. Schottky"
] | Q38104__P166 | 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 Walter Houser Brattain, the American physicist, and someone who was a student of Max Planck? | Walter Houser Brattain | Walter H. Schottky | student | Query: What are different ways of connecting Walter Houser Brattain, the American physicist, and someone who was a student of Max Planck?
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 ... | {"final_response": "<think>\n\n</think>\n\n<answer>\n{\n \"1\": [\n [\"Walter Houser Brattain\", \"co-recipient of Nobel Prize\", \"William Shockley\"],\n [\"William Shockley\", \"doctoral advisor\", \"John Bardeen\"],\n [\"John Bardeen\", \"colleague\", \"Walter H. Schottky\"],\n [\"Walter H. Schottky\", ... |
[
"Q16011894",
"P58",
"Q10363147"
] | [
"Q601629",
"P58",
"Q10363147"
] | [
"Q10363147",
"P69",
"Q49088"
] | [
"Maurice Geraghty",
"screenwriter",
"Robbers' Roost"
] | [
"Sidney Salkow",
"screenwriter",
"Robbers' Roost"
] | [
"Robbers' Roost",
"educated at",
"Columbia University"
] | Q10363147__P58 | 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 Robbers' Roost, a notable location, and someone who was educated at Columbia University? | Robbers' Roost | Columbia University | educated at | Query: What are different ways of connecting Robbers' Roost, a notable location, and someone who was educated at Columbia University?
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 cont... | {"final_response": "<think>\n\n</think>\n\n<answer>\n{\n \"1\": [\n [\n \"Robbers' Roost\",\n \"associated with\",\n \"Butch Cassidy\"\n ],\n [\n \"Butch Cassidy\",\n \"portrayed by\",\n \"Paul Newman\"\n ],\n [\n \"Paul Newman\",\n \"spouse\",\n \"Joanne Woo... |
[
"Q64019854",
"P166",
"Q17144"
] | [
"Q94041",
"P166",
"Q17144"
] | [
"Q94041",
"P136",
"Q172980"
] | [
"G. Gould Lincoln",
"award received",
"Presidential Medal of Freedom"
] | [
"James Cagney",
"award received",
"Presidential Medal of Freedom"
] | [
"James Cagney",
"genre",
"Western film"
] | Q17144__P166 | 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 G. Gould Lincoln, an author known for his works in Western literature, and someone who is associated with the genre of Western films? | G. Gould Lincoln | Western film | genre | Query: What are different ways of connecting G. Gould Lincoln, an author known for his works in Western literature, and someone who is associated with the genre of Western films?
Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual trip... | {"final_response": "<think>\n\n</think>\n\n<answer>\n{\n \"1\": [\n [\"G. Gould Lincoln\", \"wrote screenplay for\", \"The Man from Montana\"],\n [\"The Man from Montana\", \"genre\", \"Western film\"]\n ],\n \"2\": [\n [\"G. Gould Lincoln\", \"wrote screenplay for\", \"The Man from Montana\"],\n [\"The ... |
[
"Q4419898",
"P40",
"Q2494"
] | [
"Q100441774",
"P40",
"Q2494"
] | [
"Q100441774",
"P19",
"Q524660"
] | [
"Joseph Ratzinger, Sr.",
"child",
"Benedict XVI"
] | [
"Maria Peintner",
"child",
"Benedict XVI"
] | [
"Maria Peintner",
"place of birth",
"Oberaudorf"
] | Q2494__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 Joseph Ratzinger, Sr., and someone who was born in Oberaudorf? | Joseph Ratzinger, Sr. | Oberaudorf | place of birth | Query: What are different ways of connecting Joseph Ratzinger, Sr., and someone who was born in Oberaudorf?
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 w... | {"final_response": "<think>\n\n</think>\n\n<answer>\n{\n \"1\": [\n [\n \"Joseph Ratzinger, Sr.\",\n \"spouse\",\n \"Maria Ratzinger\"\n ],\n [\n \"Maria Ratzinger\",\n \"child\",\n \"Georg Ratzinger\"\n ],\n [\n \"Georg Ratzinger\",\n \"sibling\",\n \"Pope B... |
[
"Q104891260",
"P57",
"Q20675767"
] | [
"Q104891260",
"P57",
"Q18018415"
] | [
"Q18018415",
"P1344",
"Q116813843"
] | [
"The Gray Man",
"director",
"Joe Russo"
] | [
"The Gray Man",
"director",
"Anthony Russo"
] | [
"Anthony Russo",
"participant in",
"Global Conference 2019"
] | Q104891260__P57 | 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 The Gray Man, a popular film, and someone who participated in the Global Conference 2019? | The Gray Man | Global Conference 2019 | participant in | Query: What are different ways of connecting The Gray Man, a popular film, and someone who participated in the Global Conference 2019?
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 con... | {"final_response": "<think>\n\n</think>\n\n<answer>\n{\n \"1\": [\n [\"The Gray Man\", \"cast member\", \"Chris Evans\"],\n [\"Chris Evans\", \"participant in\", \"Global Conference 2019\"]\n ],\n \"2\": [\n [\"The Gray Man\", \"cast member\", \"Reg\u00e9-Jean Page\"],\n [\"Reg\u00e9-Jean Page\", \"parti... |
[
"Q1043614",
"P57",
"Q4747497"
] | [
"Q1043614",
"P57",
"Q9557"
] | [
"Q9557",
"P166",
"Q30132541"
] | [
"Taare Zameen Par",
"director",
"Amole Gupte"
] | [
"Taare Zameen Par",
"director",
"Aamir Khan"
] | [
"Aamir Khan",
"award received",
"Padma Shri in arts"
] | Q1043614__P57 | 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 Taare Zameen Par, the acclaimed film, and someone who has received the Padma Shri award in the arts? | Taare Zameen Par | Padma Shri in arts | award received | Query: What are different ways of connecting Taare Zameen Par, the acclaimed film, and someone who has received the Padma Shri award in the arts?
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) fo... | {"final_response": "<think>\n\n</think>\n\n<answer>\n{\n \"1\": [\n [\n \"Taare Zameen Par\",\n \"director\",\n \"Aamir Khan\"\n ],\n [\n \"Aamir Khan\",\n \"award received\",\n \"Padma Shri in arts\"\n ]\n ],\n \"2\": [\n [\n \"Taare Zameen Par\",\n \"screenwri... |
End of preview. Expand in Data Studio
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-pro-preview - Server URL:
Not specified - API Key:
Not provided - Request Timeout:
30 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:
low
Feature Flags
- Step by Step Reasoning:
False - Reasoning Enabled:
False - No Curator:
False - Index Filter:
None (all samples) - Start Index:
80 - End Index:
100
Data Information
- Input File:
connections-dev/connection_queries_jan12 - Number of Samples:
20 - Output Filename:
connection_queries_jan12__natural__s80__e100__original__1__low__0.7__16384__gemini-3-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("s80_e100_original_1_low_0.7_16384_gemini-3-pro-preview__raw")
Generation Details
This dataset was generated on 2026-01-22 14:55:12 using the inference pipeline with the above configuration.
- Downloads last month
- 2