Dataset Viewer
Auto-converted to Parquet Duplicate
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
11
16
keep_triple
bool
1 class
query_prompt
stringlengths
3.23k
3.3k
query
stringlengths
87
170
entity_a
stringlengths
4
53
entity_b
stringlengths
5
22
rel_b
stringlengths
5
28
prompt
stringlengths
3.55k
3.69k
index
int64
120
139
raw_responses
stringlengths
759
4.85k
[ "Q273809", "P39", "Q14211" ]
[ "Q9545", "P39", "Q14211" ]
[ "Q9545", "P69", "Q609646" ]
[ "Thomas Pelham-Holles, 1st Duke of Newcastle-upon-Tyne", "position held", "Prime Minister of the United Kingdom" ]
[ "Tony Blair", "position held", "Prime Minister of the United Kingdom" ]
[ "Tony Blair", "educated at", "St John's College" ]
Q14211__P39
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 Thomas Pelham-Holles, 1st Duke of Newcastle-upon-Tyne, and someone who was educated at St John's College?
Thomas Pelham-Holles, 1st Duke of Newcastle-upon-Tyne
St John's College
educated at
Query: What are different ways of connecting Thomas Pelham-Holles, 1st Duke of Newcastle-upon-Tyne, and someone who was educated at St John's College? 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, tai...
120
{"final_response": "```json\n{\n \"1\": [\n [\n \"Thomas Pelham-Holles, 1st Duke of Newcastle-upon-Tyne\",\n \"held office\",\n \"Prime Minister of the United Kingdom\"\n ],\n [\n \"Tony Blair\",\n \"held office\",\n \"Prime Minister of the United Kingdom\"\n ],\n [\n ...
[ "Q188293", "P166", "Q38104" ]
[ "Q78520", "P166", "Q38104" ]
[ "Q78520", "P27", "Q40" ]
[ "Frits Zernike", "award received", "Nobel Prize in Physics" ]
[ "Victor Francis Hess", "award received", "Nobel Prize in Physics" ]
[ "Victor Francis Hess", "country of citizenship", "Austria" ]
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 Frits Zernike, the Dutch physicist, and someone who holds citizenship in Austria?
Frits Zernike
Austria
country of citizenship
Query: What are different ways of connecting Frits Zernike, the Dutch physicist, and someone who holds citizenship in Austria? 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 ...
121
{"final_response": "<answer>\n{\n \"1\": [\n [\"Frits Zernike\", \"won award\", \"Nobel Prize in Physics\"],\n [\"Nobel Prize in Physics\", \"awarded to\", \"Erwin Schr\u00f6dinger\"],\n [\"Erwin Schr\u00f6dinger\", \"country of citizenship\", \"Austria\"]\n ],\n \"2\": [\n [\"Frits Zernike\", \"won awar...
[ "Q190050", "P58", "Q3178961" ]
[ "Q190050", "P58", "Q268181" ]
[ "Q268181", "P136", "Q38072107" ]
[ "Fight Club", "screenwriter", "Jim Uhls" ]
[ "Fight Club", "screenwriter", "Chuck Palahniuk" ]
[ "Chuck Palahniuk", "genre", "fiction literature" ]
Q190050__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 Jim Uhls, a screenwriter known for his adaptations, and someone who is associated with the genre of fiction literature?
Jim Uhls
fiction literature
genre
Query: What are different ways of connecting Jim Uhls, a screenwriter known for his adaptations, and someone who is associated with the genre of fiction literature? Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual triples (head, rel...
122
{"final_response": "<answer>\n{\n \"1\": [\n [\"Jim Uhls\", \"screenwriter of\", \"Fight Club (film)\"],\n [\"Fight Club (film)\", \"based on\", \"Fight Club (novel)\"],\n [\"Fight Club (novel)\", \"genre\", \"fiction literature\"]\n ],\n \"2\": [\n [\"Jim Uhls\", \"screenwriter of\", \"Jumper (film)\"],...
[ "Q2744", "P161", "Q230779" ]
[ "Q2744", "P161", "Q312077" ]
[ "Q312077", "P40", "Q112121442" ]
[ "The X-Files", "cast member", "Annabeth Gish" ]
[ "The X-Files", "cast member", "Lance Henriksen" ]
[ "Lance Henriksen", "child", "Alcamy Henriksen" ]
Q2744__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 Annabeth Gish, the American actress, and someone who is the parent of Alcamy Henriksen?
Annabeth Gish
Alcamy Henriksen
child
Query: What are different ways of connecting Annabeth Gish, the American actress, and someone who is the parent of Alcamy Henriksen? 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 conti...
123
{"final_response": "<answer>\n{\n \"1\": [\n [\"Annabeth Gish\", \"cast member\", \"The X-Files\"],\n [\"Lance Henriksen\", \"guest star\", \"The X-Files\"],\n [\"Lance Henriksen\", \"child\", \"Alcamy Henriksen\"]\n ],\n \"2\": [\n [\"Annabeth Gish\", \"cast member\", \"The X-Files\"],\n [\"The X-Fil...
[ "Q23", "P39", "Q11696" ]
[ "Q12325", "P39", "Q11696" ]
[ "Q12325", "P607", "Q26013" ]
[ "George Washington", "position held", "President of the United States" ]
[ "James Buchanan", "position held", "President of the United States" ]
[ "James Buchanan", "conflict", "War of 1812" ]
Q11696__P39
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 George Washington, the first President of the United States, and someone who was involved in the conflict known as the War of 1812?
George Washington
War of 1812
conflict
Query: What are different ways of connecting George Washington, the first President of the United States, and someone who was involved in the conflict known as the War of 1812? Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual triple...
124
{"final_response": "<answer>\n{\n \"1\": [\n [\"George Washington\", \"succeeded by\", \"John Adams\"],\n [\"John Adams\", \"succeeded by\", \"Thomas Jefferson\"],\n [\"Thomas Jefferson\", \"succeeded by\", \"James Madison\"],\n [\"James Madison\", \"conflict\", \"War of 1812\"]\n ],\n \"2\": [\n [\"G...
[ "Q107420", "P166", "Q38104" ]
[ "Q201452", "P166", "Q38104" ]
[ "Q201452", "P106", "Q752129" ]
[ "Sheldon Glashow", "award received", "Nobel Prize in Physics" ]
[ "Brian Schmidt", "award received", "Nobel Prize in Physics" ]
[ "Brian Schmidt", "occupation", "astrophysicist" ]
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 Sheldon Glashow, the physicist, and someone who is an astrophysicist?
Sheldon Glashow
astrophysicist
occupation
Query: What are different ways of connecting Sheldon Glashow, the physicist, and someone who is an astrophysicist? 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 b...
125
{"final_response": "<answer>\n{\n \"1\": [\n [\n \"Sheldon Glashow\",\n \"educated at\",\n \"Bronx High School of Science\"\n ],\n [\n \"Neil deGrasse Tyson\",\n \"educated at\",\n \"Bronx High School of Science\"\n ],\n [\n \"Neil deGrasse Tyson\",\n \"occupation...
[ "Q20817224", "P2175", "Q16501" ]
[ "Q327362", "P2175", "Q16501" ]
[ "Q327362", "P2868", "Q422106" ]
[ "thyroglobulin", "medical condition treated", "hypothyroidism" ]
[ "triiodothyronine", "medical condition treated", "hypothyroidism" ]
[ "triiodothyronine", "subject has role", "thyroid hormone" ]
Q16501__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 thyroglobulin, a protein involved in thyroid function, and a substance that plays a role as a thyroid hormone?
thyroglobulin
thyroid hormone
subject has role
Query: What are different ways of connecting thyroglobulin, a protein involved in thyroid function, and a substance that plays a role as a thyroid hormone? 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...
126
{"final_response": "<answer>\n{\n \"1\": [\n [\"thyroglobulin\", \"precursor of\", \"thyroxine\"],\n [\"thyroxine\", \"subject has role\", \"thyroid hormone\"]\n ],\n \"2\": [\n [\"thyroglobulin\", \"precursor of\", \"triiodothyronine\"],\n [\"triiodothyronine\", \"subject has role\", \"thyroid hormone\"...
[ "Q114273718", "P57", "Q1050421" ]
[ "Q16948256", "P57", "Q1050421" ]
[ "Q1050421", "P106", "Q28389" ]
[ "Katerina Kitidi", "director", "Catastroika" ]
[ "Aris Chatzistefanou", "director", "Catastroika" ]
[ "Catastroika", "occupation", "screenwriter" ]
Q1050421__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 Katerina Kitidi, a notable figure in the arts, and someone who works as a screenwriter?
Katerina Kitidi
screenwriter
occupation
Query: What are different ways of connecting Katerina Kitidi, a notable figure in the arts, and someone who works as a screenwriter? 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 conti...
127
{"final_response": "<answer>\n{\n \"1\": [\n \"Katerina Kitidi\", \"occupation\", \"screenwriter\"\n ],\n \"2\": [\n \"Katerina Kitidi\", \"directed\", \"Debtocracy\",\n \"Aris Chatzistefanou\", \"directed\", \"Debtocracy\",\n \"Aris Chatzistefanou\", \"occupation\", \"screenwriter\"\n ],\n \"3\": [\n ...
[ "Q7614131", "P54", "Q190618" ]
[ "Q4934212", "P54", "Q190618" ]
[ "Q4934212", "P69", "Q230492" ]
[ "Steve Thurlow", "member of sports team", "New York Giants" ]
[ "Bob Topp", "member of sports team", "New York Giants" ]
[ "Bob Topp", "educated at", "University of Michigan" ]
Q190618__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 Steve Thurlow and someone who was educated at the University of Michigan?
Steve Thurlow
University of Michigan
educated at
Query: What are different ways of connecting Steve Thurlow and someone who was educated at the University of Michigan? 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 th...
128
{"final_response": "<answer>\n{\n \"1\": [\n [\"Steve Thurlow\", \"played for\", \"New York Giants\"],\n [\"New York Giants\", \"drafted\", \"Amani Toomer\"],\n [\"Amani Toomer\", \"educated at\", \"University of Michigan\"]\n ],\n \"2\": [\n [\"Steve Thurlow\", \"played for\", \"Washington Redskins\"],\...
[ "Q1033108", "P162", "Q7612676" ]
[ "Q1033108", "P162", "Q461913" ]
[ "Q461913", "P19", "Q1345" ]
[ "Candyman", "producer", "Steve Golin" ]
[ "Candyman", "producer", "Alan Poul" ]
[ "Alan Poul", "place of birth", "Philadelphia" ]
Q1033108__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 Steve Golin, the renowned film producer, and someone who was born in Philadelphia?
Steve Golin
Philadelphia
place of birth
Query: What are different ways of connecting Steve Golin, the renowned film producer, and someone who was born in Philadelphia? 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...
129
{"final_response": "```json\n<answer>\n{\n \"1\": [\n [\"Steve Golin\", \"producer of\", \"Sleepers\"],\n [\"Sleepers\", \"cast member\", \"Kevin Bacon\"],\n [\"Kevin Bacon\", \"place of birth\", \"Philadelphia\"]\n ],\n \"2\": [\n [\"Steve Golin\", \"producer of\", \"Case 39\"],\n [\"Case 39\", \"cas...
[ "Q18041516", "P2293", "Q11081" ]
[ "Q18036616", "P2293", "Q11081" ]
[ "Q18036616", "P684", "Q18254158" ]
[ "ZCWPW1", "genetic association", "Alzheimer's disease" ]
[ "SLC4A1AP", "genetic association", "Alzheimer's disease" ]
[ "SLC4A1AP", "ortholog", "Slc4a1ap" ]
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 ZCWPW1, a gene, and a substance that is an ortholog of SLC4A1AP?
ZCWPW1
Slc4a1ap
ortholog
Query: What are different ways of connecting ZCWPW1, a gene, and a substance that is an ortholog of SLC4A1AP? 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...
130
{"final_response": "<answer>\n{\n \"1\": [\n [\"ZCWPW1\", \"found in taxon\", \"Homo sapiens\"],\n [\"Homo sapiens\", \"expresses gene\", \"SLC4A1AP\"],\n [\"SLC4A1AP\", \"ortholog\", \"Slc4a1ap\"]\n ],\n \"2\": [\n [\"ZCWPW1\", \"interacts with\", \"UBC\"],\n [\"UBC\", \"interacts with\", \"SLC4A1AP\...
[ "Q213425", "P166", "Q17144" ]
[ "Q553147", "P166", "Q17144" ]
[ "Q553147", "P22", "Q4980614" ]
[ "Danny Kaye", "award received", "Presidential Medal of Freedom" ]
[ "Gordon B. Hinckley", "award received", "Presidential Medal of Freedom" ]
[ "Gordon B. Hinckley", "father", "Bryant S. Hinckley" ]
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 Danny Kaye, the beloved entertainer, and someone who is the father of Bryant S. Hinckley?
Danny Kaye
Bryant S. Hinckley
father
Query: What are different ways of connecting Danny Kaye, the beloved entertainer, and someone who is the father of Bryant S. Hinckley? 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...
131
{"final_response": "<answer>\n{\n \"1\": [\n [\"Danny Kaye\", \"awarded\", \"Presidential Medal of Freedom\"],\n [\"Gordon B. Hinckley\", \"awarded\", \"Presidential Medal of Freedom\"],\n [\"Gordon B. Hinckley\", \"grandson of\", \"Ira Nathaniel Hinckley\"],\n [\"Ira Nathaniel Hinckley\", \"father\", \"Br...
[ "Q18029171", "P2293", "Q11081" ]
[ "Q17855637", "P2293", "Q11081" ]
[ "Q17855637", "P684", "Q18248084" ]
[ "AFF1", "genetic association", "Alzheimer's disease" ]
[ "CAMK4", "genetic association", "Alzheimer's disease" ]
[ "CAMK4", "ortholog", "Camk4" ]
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 AFF1, a gene associated with various biological functions, and a substance that is an ortholog of CAMK4?
AFF1
Camk4
ortholog
Query: What are different ways of connecting AFF1, a gene associated with various biological functions, and a substance that is an ortholog of CAMK4? 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...
132
{"final_response": "<answer>\n{\n \"1\": [\n [\"AFF1\", \"interacts with\", \"CREBBP\"],\n [\"CREBBP\", \"interacts with\", \"CAMK4\"],\n [\"CAMK4\", \"ortholog\", \"Camk4\"]\n ],\n \"2\": [\n [\"AFF1\", \"interacts with\", \"TP53\"],\n [\"TP53\", \"interacts with\", \"CAMK4\"],\n [\"CAMK4\", \"ort...
[ "Q34597", "P39", "Q11696" ]
[ "Q35678", "P39", "Q11696" ]
[ "Q35678", "P106", "Q189290" ]
[ "James A. Garfield", "position held", "President of the United States" ]
[ "Benjamin Harrison", "position held", "President of the United States" ]
[ "Benjamin Harrison", "occupation", "military officer" ]
Q11696__P39
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 James A. Garfield, the 20th President of the United States, and someone who has served as a military officer?
James A. Garfield
military officer
occupation
Query: What are different ways of connecting James A. Garfield, the 20th President of the United States, and someone who has served as a military officer? 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,...
133
{"final_response": "<answer>\n{\n \"1\": [\n [\"James A. Garfield\", \"occupation\", \"military officer\"]\n ],\n \"2\": [\n [\"James A. Garfield\", \"preceded by\", \"Rutherford B. Hayes\"],\n [\"Rutherford B. Hayes\", \"occupation\", \"military officer\"]\n ],\n \"3\": [\n [\"James A. Garfield\", \"s...
[ "Q422654", "P2175", "Q12199" ]
[ "Q422618", "P2175", "Q12199" ]
[ "Q422618", "P769", "Q408535" ]
[ "saquinavir", "medical condition treated", "HIV/AIDS" ]
[ "ritonavir", "medical condition treated", "HIV/AIDS" ]
[ "ritonavir", "significant drug interaction", "quetiapine" ]
Q12199__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 saquinavir, a medication used to treat HIV, and a substance that has a significant drug interaction with quetiapine?
saquinavir
quetiapine
significant drug interaction
Query: What are different ways of connecting saquinavir, a medication used to treat HIV, and a substance that has a significant drug interaction with quetiapine? 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...
134
{"final_response": "<answer>\n{\n \"1\": [\n [\"saquinavir\", \"co-administered with\", \"ritonavir\"],\n [\"ritonavir\", \"significant drug interaction\", \"quetiapine\"]\n ],\n \"2\": [\n [\"saquinavir\", \"interacts with\", \"ketoconazole\"],\n [\"ketoconazole\", \"significant drug interaction\", \"qu...
[ "Q2875", "P161", "Q1577419" ]
[ "Q2875", "P161", "Q18811009" ]
[ "Q18811009", "P19", "Q729304" ]
[ "Gone with the Wind", "cast member", "Hans Moebus" ]
[ "Gone with the Wind", "cast member", "David Newell" ]
[ "David Newell", "place of birth", "Carthage" ]
Q2875__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 Hans Moebus and someone who was born in Carthage?
Hans Moebus
Carthage
place of birth
Query: What are different ways of connecting Hans Moebus and someone who was born in Carthage? 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 one entit...
135
{"final_response": "```json\n{\n \"1\": [\n [\n \"Hans Moebus\",\n \"occupation\",\n \"Actor\"\n ],\n [\n \"Victor Mature\",\n \"occupation\",\n \"Actor\"\n ],\n [\n \"Victor Mature\",\n \"played role\",\n \"Hannibal\"\n ],\n [\n \"Hannibal\",\n ...
[ "Q20817224", "P2175", "Q16501" ]
[ "Q773449", "P2175", "Q16501" ]
[ "Q773449", "P3364", "Q5268499" ]
[ "thyroglobulin", "medical condition treated", "hypothyroidism" ]
[ "L-thyroxine", "medical condition treated", "hypothyroidism" ]
[ "L-thyroxine", "stereoisomer of", "dextrothyroxine" ]
Q16501__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 thyroglobulin, a protein involved in thyroid hormone production, and a substance that is a stereoisomer of dextrothyroxine?
thyroglobulin
dextrothyroxine
stereoisomer of
Query: What are different ways of connecting thyroglobulin, a protein involved in thyroid hormone production, and a substance that is a stereoisomer of dextrothyroxine? Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual triples (head,...
136
{"final_response": "<answer>\n{\n \"1\": [\n [\"thyroglobulin\", \"precursor of\", \"levothyroxine\"],\n [\"levothyroxine\", \"stereoisomer of\", \"dextrothyroxine\"]\n ],\n \"2\": [\n [\"thyroglobulin\", \"produced by\", \"thyroid gland\"],\n [\"thyroid gland\", \"secretes\", \"levothyroxine\"],\n [\...
[ "Q424027", "P2175", "Q12192" ]
[ "Q412236", "P2175", "Q12192" ]
[ "Q412236", "P769", "Q191521" ]
[ "vancomycin", "medical condition treated", "pneumonia" ]
[ "voriconazole", "medical condition treated", "pneumonia" ]
[ "voriconazole", "significant drug interaction", "sildenafil" ]
Q12192__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 vancomycin, an antibiotic used to treat serious infections, and a substance that has a significant drug interaction with sildenafil?
vancomycin
sildenafil
significant drug interaction
Query: What are different ways of connecting vancomycin, an antibiotic used to treat serious infections, and a substance that has a significant drug interaction with sildenafil? Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual tripl...
137
{"final_response": "<answer>\n{\n \"1\": [\n [\n \"Vancomycin\",\n \"discovered by\",\n \"Eli Lilly and Company\"\n ],\n [\n \"Eli Lilly and Company\",\n \"discovered\",\n \"Erythromycin\"\n ],\n [\n \"Erythromycin\",\n \"significant drug interaction\",\n \"S...
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-3.1-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: 120
  • End Index: 140

Data Information

  • Input File: connections-dev/connection_queries_jan12
  • Number of Samples: 20
  • Output Filename: connection_queries_jan12__natural__s120__e140__original__1__medium__0.7__16384__gemini-3_1-pro-preview.jsonl

Dataset Structure

The dataset contains the following key columns:

  • query: The original query
  • entity_a: Starting entity
  • entity_b: Target entity
  • rel_b: Target relationship
  • path_prediction: Generated connection paths
  • prompt_for_path: Prompts used for path generation

Usage

You can load this dataset using:

from datasets import load_dataset

dataset = load_dataset("s120_e140_original_1_medium_0.7_16384_gemini-3_1-pro-preview__raw")

Generation Details

This dataset was generated on 2026-03-12 01:50:09 using the inference pipeline with the above configuration.

Downloads last month
9