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 14 | keep_triple bool 1
class | query_prompt stringlengths 3.24k 3.32k | query stringlengths 88 226 | entity_a stringlengths 4 25 | entity_b stringlengths 4 60 | rel_b stringlengths 6 36 | prompt stringlengths 3.54k 3.75k | index int64 400 419 | raw_responses stringlengths 1.21k 2.58k |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
[
"Q66102",
"P463",
"Q30773"
] | [
"Q75431",
"P463",
"Q30773"
] | [
"Q75431",
"P19",
"Q1794"
] | [
"Mandy Capristo",
"member of",
"Monrose"
] | [
"Senna Guemmour",
"member of",
"Monrose"
] | [
"Senna Guemmour",
"place of birth",
"Frankfurt"
] | Q30773__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 Mandy Capristo, the German singer, and someone who was born in Frankfurt? | Mandy Capristo | Frankfurt | place of birth | Query: What are different ways of connecting Mandy Capristo, the German singer, and someone who was born in Frankfurt?
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... | 400 | {"final_response": "<answer>\n{\n \"1\": [\n [\"Mandy Capristo\", \"bandmate of\", \"Senna Gammour\"],\n [\"Senna Gammour\", \"place of birth\", \"Frankfurt\"]\n ],\n \"2\": [\n [\"Mandy Capristo\", \"member of\", \"Monrose\"],\n [\"Monrose\", \"has member\", \"Senna Gammour\"],\n [\"Senna Gammour\", ... |
[
"Q714",
"P166",
"Q935843"
] | [
"Q216924",
"P166",
"Q935843"
] | [
"Q216924",
"P1344",
"Q1233991"
] | [
"Stevie Wonder",
"award received",
"Grammy Lifetime Achievement Award"
] | [
"Glenn Gould",
"award received",
"Grammy Lifetime Achievement Award"
] | [
"Glenn Gould",
"participant in",
"documenta 8"
] | Q935843__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 Stevie Wonder, the legendary musician, and someone who participated in Documenta 8? | Stevie Wonder | documenta 8 | participant in | Query: What are different ways of connecting Stevie Wonder, the legendary musician, and someone who participated in Documenta 8?
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 continuou... | 401 | {"final_response": "```json\n<answer>\n{\n \"1\": [\n [\"Stevie Wonder\", \"collaborated with\", \"Bruce Springsteen\"],\n [\"Bruce Springsteen\", \"collaborated with\", \"Lou Reed\"],\n [\"Lou Reed\", \"spouse of\", \"Laurie Anderson\"],\n [\"Laurie Anderson\", \"participant in\", \"documenta 8\"]\n ],\n... |
[
"Q347337",
"P2175",
"Q1477"
] | [
"Q416132",
"P2175",
"Q1477"
] | [
"Q416132",
"P769",
"Q424940"
] | [
"balsalazide",
"medical condition treated",
"ulcerative colitis"
] | [
"betamethasone",
"medical condition treated",
"ulcerative colitis"
] | [
"betamethasone",
"significant drug interaction",
"moxifloxacin"
] | Q1477__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 balsalazide, a medication used to treat inflammatory bowel disease, and a substance that has a significant drug interaction with moxifloxacin? | balsalazide | moxifloxacin | significant drug interaction | Query: What are different ways of connecting balsalazide, a medication used to treat inflammatory bowel disease, and a substance that has a significant drug interaction with moxifloxacin?
Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of fac... | 402 | {"final_response": "<answer>\n{\n \"1\": [\n [\"balsalazide\", \"treats\", \"ulcerative colitis\"],\n [\"prednisone\", \"treats\", \"ulcerative colitis\"],\n [\"prednisone\", \"significant drug interaction\", \"moxifloxacin\"]\n ],\n \"2\": [\n [\"balsalazide\", \"causes side effect\", \"dyspepsia\"],\n ... |
[
"Q484048",
"P161",
"Q491766"
] | [
"Q484048",
"P161",
"Q2965745"
] | [
"Q2965745",
"P26",
"Q3171644"
] | [
"Amélie",
"cast member",
"Mathieu Kassovitz"
] | [
"Amélie",
"cast member",
"Christiane Bopp"
] | [
"Christiane Bopp",
"spouse",
"Jean Dell"
] | Q484048__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 Mathieu Kassovitz, the French filmmaker and actor, and someone who is the spouse of Jean Dell? | Mathieu Kassovitz | Jean Dell | spouse | Query: What are different ways of connecting Mathieu Kassovitz, the French filmmaker and actor, and someone who is the spouse of Jean Dell?
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 ... | 403 | {"final_response": "```json\n{\n \"1\": [\n [\n \"Mathieu Kassovitz\",\n \"country of citizenship\",\n \"France\"\n ],\n [\n \"Christiane Bopp\",\n \"country of citizenship\",\n \"France\"\n ],\n [\n \"Christiane Bopp\",\n \"spouse\",\n \"Jean Dell\"\n ]\n... |
[
"Q6388049",
"P54",
"Q337758"
] | [
"Q7846877",
"P54",
"Q337758"
] | [
"Q7846877",
"P69",
"Q7401790"
] | [
"Ken MacAfee",
"member of sports team",
"San Francisco 49ers"
] | [
"Troy Kyles",
"member of sports team",
"San Francisco 49ers"
] | [
"Troy Kyles",
"educated at",
"Saint Martin de Porres High School"
] | 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 Ken MacAfee, a notable figure, and someone who was educated at Saint Martin de Porres High School? | Ken MacAfee | Saint Martin de Porres High School | educated at | Query: What are different ways of connecting Ken MacAfee, a notable figure, and someone who was educated at Saint Martin de Porres High 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) form... | 404 | {"final_response": "<answer>\n{\n \"1\": [\n [\n \"Ken MacAfee\",\n \"played for\",\n \"San Francisco 49ers\"\n ],\n [\n \"San Francisco 49ers\",\n \"played against\",\n \"Detroit Lions\"\n ],\n [\n \"Ron Rice\",\n \"played for\",\n \"Detroit Lions\"\n ],\... |
[
"Q83331",
"P166",
"Q38104"
] | [
"Q184577",
"P166",
"Q38104"
] | [
"Q184577",
"P26",
"Q109333210"
] | [
"Louis de Broglie",
"award received",
"Nobel Prize in Physics"
] | [
"Walter Houser Brattain",
"award received",
"Nobel Prize in Physics"
] | [
"Walter Houser Brattain",
"spouse",
"Keren Gilmore Brattain"
] | 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 Louis de Broglie, the French physicist, and someone who is the spouse of Walter Houser Brattain? | Louis de Broglie | Keren Gilmore Brattain | spouse | Query: What are different ways of connecting Louis de Broglie, the French physicist, and someone who is the spouse of Walter Houser Brattain?
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... | 405 | {"final_response": "<answer>\n{\n \"1\": [\n [\"Louis de Broglie\", \"awarded\", \"Nobel Prize in Physics\"],\n [\"Nobel Prize in Physics\", \"awarded to\", \"Walter Houser Brattain\"],\n [\"Walter Houser Brattain\", \"spouse\", \"Keren Gilmore Brattain\"]\n ],\n \"2\": [\n [\"Louis de Broglie\", \"formu... |
[
"Q7631165",
"P40",
"Q1316"
] | [
"Q16051068",
"P40",
"Q1316"
] | [
"Q16051068",
"P119",
"Q796"
] | [
"Subha Tulfah al-Mussallat",
"child",
"Saddam Hussein"
] | [
"Hussein 'Abid al-Majid",
"child",
"Saddam Hussein"
] | [
"Hussein 'Abid al-Majid",
"place of burial",
"Iraq"
] | Q1316__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 Subha Tulfah al-Mussallat and someone who is buried in Iraq? | Subha Tulfah al-Mussallat | Iraq | place of burial | Query: What are different ways of connecting Subha Tulfah al-Mussallat and someone who is buried in Iraq?
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 wit... | 406 | {"final_response": "<answer>\n{\n \"1\": [\n [\"Subha Tulfah al-Mussallat\", \"mother of\", \"Saddam Hussein\"],\n [\"Saddam Hussein\", \"place of burial\", \"Iraq\"]\n ],\n \"2\": [\n [\"Subha Tulfah al-Mussallat\", \"mother of\", \"Barzan Ibrahim al-Tikriti\"],\n [\"Barzan Ibrahim al-Tikriti\", \"place... |
[
"Q2303035",
"P40",
"Q913"
] | [
"Q2263186",
"P40",
"Q913"
] | [
"Q2263186",
"P106",
"Q185196"
] | [
"Sophroniscus",
"child",
"Socrates"
] | [
"Phaenarete",
"child",
"Socrates"
] | [
"Phaenarete",
"occupation",
"midwife"
] | Q913__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 Sophroniscus, a historical figure, and someone who is known for their occupation as a midwife? | Sophroniscus | midwife | occupation | Query: What are different ways of connecting Sophroniscus, a historical figure, and someone who is known for their occupation as a midwife?
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 ... | 407 | {"final_response": "<answer>\n{\n \"1\": [\n [\"Sophroniscus\", \"spouse\", \"Phaenarete\"],\n [\"Phaenarete\", \"occupation\", \"midwife\"]\n ],\n \"2\": [\n [\"Sophroniscus\", \"father of\", \"Socrates\"],\n [\"Socrates\", \"mother is\", \"Phaenarete\"],\n [\"Phaenarete\", \"occupation\", \"midwife\... |
[
"Q18049313",
"P2293",
"Q52849"
] | [
"Q18036682",
"P2293",
"Q52849"
] | [
"Q18036682",
"P5572",
"Q2603645"
] | [
"ANTXR2",
"genetic association",
"ankylosing spondylitis"
] | [
"KIF21B",
"genetic association",
"ankylosing spondylitis"
] | [
"KIF21B",
"expressed in",
"entorhinal cortex"
] | Q52849__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 ANTXR2, a protein, and a substance that is expressed in the entorhinal cortex? | ANTXR2 | entorhinal cortex | expressed in | Query: What are different ways of connecting ANTXR2, a protein, and a substance that is expressed in the entorhinal cortex?
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 cha... | 408 | {"final_response": "<answer>\n{\n \"1\": [\n (\"ANTXR2\", \"interacts with\", \"LRP6\"),\n (\"LRP6\", \"interacts with\", \"APOE\"),\n (\"APOE\", \"expressed in\", \"entorhinal cortex\")\n ],\n \"2\": [\n (\"ANTXR2\", \"interacts with\", \"FN1\"),\n (\"FN1\", \"expressed in\", \"entorhinal cortex\")\n... |
[
"Q277785",
"P2175",
"Q41861"
] | [
"Q414873",
"P2175",
"Q41861"
] | [
"Q414873",
"P129",
"Q14905748"
] | [
"perindopril",
"medical condition treated",
"arterial hypertension"
] | [
"isradipine",
"medical condition treated",
"arterial hypertension"
] | [
"isradipine",
"physically interacts with",
"Calcium channel, voltage-dependent, L type, alpha 1C subunit"
] | 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 perindopril, a medication used to treat high blood pressure, and a substance that physically interacts with the voltage-dependent L-type calcium channel, specifically the alpha 1C subunit? | perindopril | Calcium channel, voltage-dependent, L type, alpha 1C subunit | physically interacts with | Query: What are different ways of connecting perindopril, a medication used to treat high blood pressure, and a substance that physically interacts with the voltage-dependent L-type calcium channel, specifically the alpha 1C subunit?
Task: Identify how two real-world entities are connected by producing MANY connection... | 409 | {"final_response": "<answer>\n{\n \"1\": [\n [\"perindopril\", \"formulated with\", \"amlodipine\"],\n [\"amlodipine\", \"physically interacts with\", \"Calcium channel, voltage-dependent, L type, alpha 1C subunit\"]\n ],\n \"2\": [\n [\"perindopril\", \"treats\", \"hypertension\"],\n [\"hypertension\", ... |
[
"Q4044993",
"P2293",
"Q131755"
] | [
"Q17927578",
"P2293",
"Q131755"
] | [
"Q17927578",
"P5572",
"Q193302"
] | [
"NPAS3",
"genetic association",
"bipolar disorder"
] | [
"FBLN1",
"genetic association",
"bipolar disorder"
] | [
"FBLN1",
"expressed in",
"pericardium"
] | 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 NPAS3, a gene associated with various biological functions, and a substance that is expressed in the pericardium? | NPAS3 | pericardium | expressed in | Query: What are different ways of connecting NPAS3, a gene associated with various biological functions, and a substance that is expressed in the pericardium?
Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual triples (head, relations... | 410 | {"final_response": "<answer>\n{\n \"1\": [\n [\"NPAS3\", \"interacts with\", \"ARNT\"],\n [\"ARNT\", \"expressed in\", \"pericardium\"]\n ],\n \"2\": [\n [\"NPAS3\", \"interacts with\", \"ARNTL\"],\n [\"ARNTL\", \"expressed in\", \"pericardium\"]\n ],\n \"3\": [\n [\"NPAS3\", \"interacts with\", \"E... |
[
"Q22918544",
"P463",
"Q35918"
] | [
"Q20666484",
"P463",
"Q35918"
] | [
"Q20666484",
"P106",
"Q50279140"
] | [
"Aina Vilberh",
"member of",
"Nu Virgos"
] | [
"Dasha Rostova",
"member of",
"Nu Virgos"
] | [
"Dasha Rostova",
"occupation",
"Twitch streamer"
] | Q35918__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 Aina Vilberh and someone who is a Twitch streamer? | Aina Vilberh | Twitch streamer | occupation | Query: What are different ways of connecting Aina Vilberh and someone who is a Twitch streamer?
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 enti... | 411 | {"final_response": "<answer>\n{\n \"1\": [\n [\"Aina Vilberh\", \"country of citizenship\", \"Ukraine\"],\n [\"Oleksandr Kostyliev\", \"country of citizenship\", \"Ukraine\"],\n [\"Oleksandr Kostyliev\", \"occupation\", \"Twitch streamer\"]\n ],\n \"2\": [\n [\"Aina Vilberh\", \"place of birth\", \"Kyiv\... |
[
"Q10363147",
"P58",
"Q16011894"
] | [
"Q10363147",
"P58",
"Q146763"
] | [
"Q146763",
"P551",
"Q8066104"
] | [
"Robbers' Roost",
"screenwriter",
"Maurice Geraghty"
] | [
"Robbers' Roost",
"screenwriter",
"Zane Grey"
] | [
"Zane Grey",
"residence",
"Zane Grey Museum"
] | 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 Maurice Geraghty, a notable figure, and someone who has a residence that is now the Zane Grey Museum? | Maurice Geraghty | Zane Grey Museum | residence | Query: What are different ways of connecting Maurice Geraghty, a notable figure, and someone who has a residence that is now the Zane Grey Museum?
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... | 412 | {"final_response": "<answer>\n{\n \"1\": [\n [\"Maurice Geraghty\", \"produced\", \"Nevada (1944 film)\"],\n [\"Nevada (1944 film)\", \"based on work by\", \"Zane Grey\"],\n [\"Zane Grey\", \"residence\", \"Zane Grey Museum\"]\n ],\n \"2\": [\n [\"Maurice Geraghty\", \"produced\", \"West of the Pecos (19... |
[
"Q14905744",
"P2293",
"Q131755"
] | [
"Q18040277",
"P2293",
"Q131755"
] | [
"Q18040277",
"P5572",
"Q744391"
] | [
"CACNA1C",
"genetic association",
"bipolar disorder"
] | [
"PRKAG2",
"genetic association",
"bipolar disorder"
] | [
"PRKAG2",
"expressed in",
"left coronary artery"
] | 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 CACNA1C, a gene associated with calcium channels, and a substance that is expressed in the left coronary artery? | CACNA1C | left coronary artery | expressed in | Query: What are different ways of connecting CACNA1C, a gene associated with calcium channels, and a substance that is expressed in the left coronary artery?
Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual triples (head, relationsh... | 413 | {"final_response": "<answer>\n{\n \"1\": [\n [\"CACNA1C\", \"expressed in\", \"left coronary artery\"]\n ],\n \"2\": [\n [\"CACNA1C\", \"interacts with\", \"CALM1\"],\n [\"CALM1\", \"expressed in\", \"left coronary artery\"]\n ],\n \"3\": [\n [\"CACNA1C\", \"interacts with\", \"RYR2\"],\n [\"RYR2\",... |
[
"Q521829",
"P54",
"Q169138"
] | [
"Q2332594",
"P54",
"Q169138"
] | [
"Q2332594",
"P413",
"Q222052"
] | [
"Antoine Walker",
"member of sports team",
"Miami Heat"
] | [
"Scott Hastings",
"member of sports team",
"Miami Heat"
] | [
"Scott Hastings",
"position played on team / speciality",
"center"
] | Q169138__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 Antoine Walker, the former professional basketball player, and someone who played the center position on a basketball team? | Antoine Walker | center | position played on team / speciality | Query: What are different ways of connecting Antoine Walker, the former professional basketball player, and someone who played the center position on a basketball 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,... | 414 | {"final_response": "<answer>\n{\n \"1\": [\n [\"Antoine Walker\", \"teammate\", \"Shaquille O'Neal\"],\n [\"Shaquille O'Neal\", \"position played\", \"center\"]\n ],\n \"2\": [\n [\"Antoine Walker\", \"teammate\", \"Kendrick Perkins\"],\n [\"Kendrick Perkins\", \"position played\", \"center\"]\n ],\n \... |
[
"Q1430",
"P40",
"Q1434"
] | [
"Q236936",
"P40",
"Q1434"
] | [
"Q236936",
"P451",
"Q1529767"
] | [
"Marcus Aurelius",
"child",
"Commodus"
] | [
"Faustina the Younger",
"child",
"Commodus"
] | [
"Faustina the Younger",
"unmarried partner",
"Lucius Tutilius Pontianus Gentianus"
] | Q1434__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 Marcus Aurelius, the Roman Emperor, and someone who was an unmarried partner of Faustina the Younger? | Marcus Aurelius | Lucius Tutilius Pontianus Gentianus | unmarried partner | Query: What are different ways of connecting Marcus Aurelius, the Roman Emperor, and someone who was an unmarried partner of Faustina the Younger?
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... | 415 | {"final_response": "<answer>\n{\n \"1\": [\n [\n \"Marcus Aurelius\",\n \"spouse\",\n \"Faustina the Younger\"\n ],\n [\n \"Faustina the Younger\",\n \"unmarried partner\",\n \"Lucius Tutilius Pontianus Gentianus\"\n ]\n ],\n \"2\": [\n [\n \"Marcus Aurelius\",\n ... |
[
"Q18047913",
"P2293",
"Q131755"
] | [
"Q18040268",
"P2293",
"Q131755"
] | [
"Q18040268",
"P5572",
"Q213456"
] | [
"CGNL1",
"genetic association",
"bipolar disorder"
] | [
"COMMD10",
"genetic association",
"bipolar disorder"
] | [
"COMMD10",
"expressed in",
"gonad"
] | 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 CGNL1, a gene, and a substance that is expressed in the gonad? | CGNL1 | gonad | expressed in | Query: What are different ways of connecting CGNL1, a gene, and a substance that is expressed in the gonad?
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... | 416 | {"final_response": "<answer>\n{\n \"1\": [\n [\"CGNL1\", \"interacts with\", \"TJP1\"],\n [\"TJP1\", \"expressed in\", \"gonad\"]\n ],\n \"2\": [\n [\"CGNL1\", \"regulates\", \"RHOA\"],\n [\"RHOA\", \"expressed in\", \"gonad\"]\n ],\n \"3\": [\n [\"CGNL1\", \"regulates\", \"RAC1\"],\n [\"RAC1\", ... |
[
"Q18034267",
"P2293",
"Q133823"
] | [
"Q18045527",
"P2293",
"Q133823"
] | [
"Q18045527",
"P5572",
"Q28455175"
] | [
"FHL5",
"genetic association",
"migraine"
] | [
"PRDM16",
"genetic association",
"migraine"
] | [
"PRDM16",
"expressed in",
"ventricular zone"
] | Q133823__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 FHL5, a protein involved in various cellular processes, and a substance that is expressed in the ventricular zone? | FHL5 | ventricular zone | expressed in | Query: What are different ways of connecting FHL5, a protein involved in various cellular processes, and a substance that is expressed in the ventricular zone?
Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of factual triples (head, relation... | 417 | {"final_response": "```json\n{\n \"1\": [\n [\"FHL5\", \"interacts with\", \"CREM\"],\n [\"CREM\", \"interacts with\", \"CREBBP\"],\n [\"CREBBP\", \"interacts with\", \"SOX2\"],\n [\"SOX2\", \"expressed in\", \"ventricular zone\"]\n ],\n \"2\": [\n [\"FHL5\", \"interacts with\", \"CREB1\"],\n [\"CR... |
[
"Q423308",
"P2175",
"Q11085"
] | [
"Q221361",
"P2175",
"Q11085"
] | [
"Q221361",
"P769",
"Q421937"
] | [
"cabergoline",
"medical condition treated",
"Parkinson's disease"
] | [
"clozapine",
"medical condition treated",
"Parkinson's disease"
] | [
"clozapine",
"significant drug interaction",
"(RS)-hydroxyzine"
] | Q11085__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 cabergoline, a medication used to treat certain hormonal disorders, and a substance that has a significant drug interaction with (RS)-hydroxyzine? | cabergoline | (RS)-hydroxyzine | significant drug interaction | Query: What are different ways of connecting cabergoline, a medication used to treat certain hormonal disorders, and a substance that has a significant drug interaction with (RS)-hydroxyzine?
Task: Identify how two real-world entities are connected by producing MANY connection paths. A connection path is a sequence of... | 418 | {"final_response": "<answer>\n{\n \"1\": [\n [\n \"cabergoline\",\n \"interacts with\",\n \"haloperidol\"\n ],\n [\n \"haloperidol\",\n \"significant drug interaction\",\n \"(RS)-hydroxyzine\"\n ]\n ],\n \"2\": [\n [\n \"cabergoline\",\n \"interacts with\",\n ... |
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:
400 - End Index:
420
Data Information
- Input File:
connections-dev/connection_queries_jan12 - Number of Samples:
20 - Output Filename:
connection_queries_jan12__natural__s400__e420__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("s400_e420_original_1_medium_0.7_16384_gemini-3_1-pro-preview__raw")
Generation Details
This dataset was generated on 2026-03-12 05:51:27 using the inference pipeline with the above configuration.
- Downloads last month
- 10