id
stringlengths 1
12
| question
stringlengths 22
566
| golden_answers
dict | data_source
stringclasses 2
values | prompt
listlengths 1
1
| ability
stringclasses 6
values | reward_model
dict | extra_info
dict | metadata
dict |
|---|---|---|---|---|---|---|---|---|
67061
|
Beau Brummell, Lord Alvanley, Henry Midmay, And Henry Pierrepont (dubbed "the Dandy Club" by Byron) jointly hosted a masquerade ball at Watier's private club in July 1813–an event that marked Beau Brummell's downfall.
|
{
"evidence": [
"Beau Brummell_sentence_41",
"Beau Brummell_sentence_40",
"Beau Brummell_sentence_42",
"Beau Brummell_sentence_45"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBeau Brummell, Lord Alvanley, Henry Midmay, And Henry Pierrepont (dubbed \"the Dandy Club\" by Byron) jointly hosted a masquerade ball at Watier's private club in July 1813–an event that marked Beau Brummell's downfall.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Beau Brummell_sentence_41",
"Beau Brummell_sentence_40",
"Beau Brummell_sentence_42",
"Beau Brummell_sentence_45"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6184,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.412",
"value": "?search="
},
{
"operation": "search",
"time": "19.785",
"value": "The Dandy Club"
},
{
"operation": "Now on",
"time": "22.313",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "26.604",
"value": "Beau Brummell"
},
{
"operation": "Now on",
"time": "39.947",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "43.95",
"value": "In the arts Beau Brummell_section_6"
},
{
"operation": "Now on",
"time": "44.608",
"value": "Beau Brummell#In the arts Beau Brummell section 6"
},
{
"operation": "Highlighting",
"time": "128.177",
"value": "Beau Brummell_sentence_41"
},
{
"operation": "Highlighting",
"time": "134.184",
"value": "Beau Brummell_sentence_40"
},
{
"operation": "Highlighting",
"time": "213.757",
"value": "Beau Brummell_sentence_42"
},
{
"operation": "Highlighting",
"time": "217.632",
"value": "Beau Brummell_sentence_45"
},
{
"operation": "finish",
"time": "253.964",
"value": "finish"
}
]
}
|
exfever_3455
|
Judd Apatow is a director who developed an Asian sitcom that aired on Fox during the 2001–02 season.
|
{
"evidence": [
"Judd Apatow",
"Undeclared"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJudd Apatow is a director who developed an Asian sitcom that aired on Fox during the 2001–02 season.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Judd Apatow",
"Undeclared"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6185,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3307
|
James Garner starred in several television series, including The Rockford Files, which aired on an American English-language commercial broadcast television and radio network, headquartered at 30 Rockefeller Plaza in New York City, with additional major offices at a rhombic skyscraper in Universal City, California, near Los Angeles.
|
{
"evidence": [
"James Garner",
"NBC",
"10 Universal City Plaza"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJames Garner starred in several television series, including The Rockford Files, which aired on an American English-language commercial broadcast television and radio network, headquartered at 30 Rockefeller Plaza in New York City, with additional major offices at a rhombic skyscraper in Universal City, California, near Los Angeles.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"James Garner",
"NBC",
"10 Universal City Plaza"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6186,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
60182
|
Polypodium is derived from Ancient Greek, they are commonly called polypodies and are terrestrial thus grow on land.
|
{
"evidence": [
"Polypodium_sentence_3",
"Polypodium_sentence_4",
"Polypodium_sentence_5",
"Terrestrial plant_sentence_1"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nPolypodium is derived from Ancient Greek, they are commonly called polypodies and are terrestrial thus grow on land.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Polypodium_sentence_3",
"Polypodium_sentence_4",
"Polypodium_sentence_5",
"Terrestrial plant_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6187,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.862",
"value": "?search="
},
{
"operation": "search",
"time": "9.567",
"value": "Polypodium"
},
{
"operation": "Now on",
"time": "10.454",
"value": "Polypodium"
},
{
"operation": "Page search",
"time": "28.571",
"value": "Greek"
},
{
"operation": "Highlighting",
"time": "32.133",
"value": "Polypodium_sentence_3"
},
{
"operation": "Highlighting",
"time": "36.938",
"value": "Polypodium_sentence_4"
},
{
"operation": "Highlighting",
"time": "41.922",
"value": "Polypodium_sentence_5"
},
{
"operation": "hyperlink",
"time": "43.588",
"value": "terrestrial"
},
{
"operation": "Now on",
"time": "44.238",
"value": "Terrestrial plant"
},
{
"operation": "Highlighting",
"time": "49.836",
"value": "Terrestrial plant_sentence_1"
},
{
"operation": "finish",
"time": "73.203",
"value": "finish"
}
]
}
|
exfever_8170
|
Dunkirk is a film with a genre concerned with warfare, typically about naval, air, or land battles, with combat scenes central to the drama.
|
{
"evidence": [
"Dunkirk (2017 film)",
"War film"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nDunkirk is a film with a genre concerned with warfare, typically about naval, air, or land battles, with combat scenes central to the drama.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Dunkirk (2017 film)",
"War film"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6188,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8868
|
"7 Things" is a song by an American singer, songwriter, actress, and television personality who is known for her distinctive raspy voice.
|
{
"evidence": [
"7 Things",
"Miley Cyrus"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\n\"7 Things\" is a song by an American singer, songwriter, actress, and television personality who is known for her distinctive raspy voice.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"7 Things",
"Miley Cyrus"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6189,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7077
|
The Hateful Eight is a 2015 Canadian film starring an American actor who appeared in Justified and The Unicorn.
|
{
"evidence": [
"The Hateful Eight",
"Walton Goggins"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Hateful Eight is a 2015 Canadian film starring an American actor who appeared in Justified and The Unicorn.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Hateful Eight",
"Walton Goggins"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6190,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3422
|
Thomas William Hiddleston is an English actor who appeared in a 2011 American superhero film produced by an American film and television production company that is a subsidiary of Walt Disney Studios, a division of the Walt Disney Company.
|
{
"evidence": [
"Tom Hiddleston",
"Thor (film)",
"Marvel Studios"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThomas William Hiddleston is an English actor who appeared in a 2011 American superhero film produced by an American film and television production company that is a subsidiary of Walt Disney Studios, a division of the Walt Disney Company.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tom Hiddleston",
"Thor (film)",
"Marvel Studios"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6191,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
39675
|
Avia Świdnik is currently part of III liga, group IV, the fourth tier in the Polish football league.
|
{
"evidence": [
"Avia Świdnik_sentence_1",
"III liga, group IV_sentence_0"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nAvia Świdnik is currently part of III liga, group IV, the fourth tier in the Polish football league.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Avia Świdnik_sentence_1",
"III liga, group IV_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6192,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.663",
"value": "?search="
},
{
"operation": "search",
"time": "13.021",
"value": "Avia Świdnik"
},
{
"operation": "Now on",
"time": "13.721",
"value": "Avia Świdnik"
},
{
"operation": "Highlighting",
"time": "26.591",
"value": "Avia Świdnik_sentence_1"
},
{
"operation": "hyperlink",
"time": "29.419",
"value": "III liga, group IV"
},
{
"operation": "Now on",
"time": "30.408",
"value": "III liga, group IV"
},
{
"operation": "Highlighting",
"time": "74.554",
"value": "III liga, group IV_sentence_0"
},
{
"operation": "finish",
"time": "177.119",
"value": "finish"
}
]
}
|
exfever_6705
|
Shailene Diann Woodley is an American actress who made her film debut in a 2011 American comedy-drama film, starring an American actor and filmmaker who received numerous accolades, including a British Academy Film Award.
|
{
"evidence": [
"Shailene Woodley",
"The Descendants",
"George Clooney"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nShailene Diann Woodley is an American actress who made her film debut in a 2011 American comedy-drama film, starring an American actor and filmmaker who received numerous accolades, including a British Academy Film Award.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Shailene Woodley",
"The Descendants",
"George Clooney"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6193,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8353
|
Honest is the second studio album by American rapper Future that features guest appearances from an American rapper who became lead vocalist of the band N.E.R.D., which he formed with an American drummer whose role in the band is often questioned.
|
{
"evidence": [
"Honest (Future album)",
"Pharrell Williams",
"Shay Haley"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nHonest is the second studio album by American rapper Future that features guest appearances from an American rapper who became lead vocalist of the band N.E.R.D., which he formed with an American drummer whose role in the band is often questioned.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Honest (Future album)",
"Pharrell Williams",
"Shay Haley"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6194,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
68625
|
A pulse oximetry AHI rating of <5 is considered normal, 5-15 is mild and 15-30 is moderate, while anything over 30 is severe.
|
{
"evidence": [
"Pulse oximetry_sentence_10"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nA pulse oximetry AHI rating of <5 is considered normal, 5-15 is mild and 15-30 is moderate, while anything over 30 is severe.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Pulse oximetry_sentence_10"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6195,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.888",
"value": "?search="
},
{
"operation": "search",
"time": "17.954",
"value": "<5 is considered normal,"
},
{
"operation": "Now on",
"time": "21.92",
"value": "undefined"
},
{
"operation": "search",
"time": "50.754",
"value": "A pulse oximetry AHI rating of"
},
{
"operation": "Now on",
"time": "52.093",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "66.937",
"value": "a pulse oximeter ahi racing of"
},
{
"operation": "Now on",
"time": "67.803",
"value": "undefined"
},
{
"operation": "search",
"time": "118.941",
"value": "Pulse oximeter"
},
{
"operation": "Now on",
"time": "137.733",
"value": "Pulse oximeter"
},
{
"operation": "search",
"time": "203.491",
"value": "ahi rating"
},
{
"operation": "Now on",
"time": "205.578",
"value": "undefined"
},
{
"operation": "search",
"time": "216.08",
"value": "AHI"
},
{
"operation": "Now on",
"time": "217.063",
"value": "Ahi"
},
{
"operation": "search",
"time": "246.458",
"value": "while anything over 30 is severe"
},
{
"operation": "Now on",
"time": "249.809",
"value": "undefined"
},
{
"operation": "search",
"time": "282.42",
"value": "Pulse oximetry"
},
{
"operation": "Now on",
"time": "283.696",
"value": "Pulse oximetry"
},
{
"operation": "Highlighting",
"time": "334.03",
"value": "Pulse oximetry_sentence_10"
},
{
"operation": "finish",
"time": "341.204",
"value": "finish"
}
]
}
|
exfever_8037
|
The District of North Vancouver is also featured in The X-Files series.
|
{
"evidence": [
"North Vancouver (district municipality)",
"District municipality"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe District of North Vancouver is also featured in The X-Files series.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"North Vancouver (district municipality)",
"District municipality"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6196,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3161
|
Frederick Christ Trump Sr. is the richest real estate developer and businessman in America.
|
{
"evidence": [
"Fred Trump",
"The Trump Organization"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nFrederick Christ Trump Sr. is the richest real estate developer and businessman in America.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Fred Trump",
"The Trump Organization"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6197,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6483
|
Courtney Michelle Love is an American singer, songwriter, and actress married to an American singer, songwriter, and lead vocalist of the rock band Nirvana.
|
{
"evidence": [
"Courtney Love",
"Kurt Cobain"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nCourtney Michelle Love is an American singer, songwriter, and actress married to an American singer, songwriter, and lead vocalist of the rock band Nirvana.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Courtney Love",
"Kurt Cobain"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6198,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5690
|
Thirteen is a 2003 American teen drama film starring an American actress who appeared in independent films, including a 2005 American black comedy film directed by Marcos Siega.
|
{
"evidence": [
"Thirteen (2003 film)",
"Evan Rachel Wood",
"Pretty Persuasion"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThirteen is a 2003 American teen drama film starring an American actress who appeared in independent films, including a 2005 American black comedy film directed by Marcos Siega.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Thirteen (2003 film)",
"Evan Rachel Wood",
"Pretty Persuasion"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6199,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
65992
|
The 2007–08 Polish Volleyball League had no representatives for any of the top-tier official competitions for men's volleyball clubs of Europe.
|
{
"evidence": [
"2007–08 Polish Volleyball League_sentence_2"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe 2007–08 Polish Volleyball League had no representatives for any of the top-tier official competitions for men's volleyball clubs of Europe.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"2007–08 Polish Volleyball League_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6200,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.082",
"value": "undefined"
},
{
"operation": "search",
"time": "61.744",
"value": "2007–08 Polish Volleyball League"
},
{
"operation": "Now on",
"time": "62.767",
"value": "2007–08 Polish Volleyball League"
},
{
"operation": "hyperlink",
"time": "129.081",
"value": "PGE Skra Bełchatów"
},
{
"operation": "Now on",
"time": "144.878",
"value": "PGE Skra Bełchatów"
},
{
"operation": "back-button-clicked",
"time": "155.273",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "155.346",
"value": "2007–08 Polish Volleyball League"
},
{
"operation": "Highlighting",
"time": "157.222",
"value": "2007–08 Polish Volleyball League_sentence_2"
},
{
"operation": "Highlighting",
"time": "165.284",
"value": "2007–08 Polish Volleyball League_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "169.636",
"value": "2007–08 Polish Volleyball League_sentence_3"
},
{
"operation": "hyperlink",
"time": "171.199",
"value": "CEV Challenge Cup"
},
{
"operation": "Now on",
"time": "172.127",
"value": "CEV Challenge Cup"
},
{
"operation": "back-button-clicked",
"time": "178.137",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "178.209",
"value": "2007–08 Polish Volleyball League"
},
{
"operation": "hyperlink",
"time": "180.09",
"value": "Asseco Resovia Rzeszów"
},
{
"operation": "Now on",
"time": "188.194",
"value": "Resovia (volleyball)"
},
{
"operation": "back-button-clicked",
"time": "189.384",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "189.452",
"value": "2007–08 Polish Volleyball League"
},
{
"operation": "hyperlink",
"time": "192.129",
"value": "CEV Cup"
},
{
"operation": "Now on",
"time": "193.21",
"value": "CEV Cup"
},
{
"operation": "back-button-clicked",
"time": "199.382",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "199.455",
"value": "2007–08 Polish Volleyball League"
},
{
"operation": "hyperlink",
"time": "204.327",
"value": "Mlekpol AZS Olsztyn"
},
{
"operation": "Now on",
"time": "206.707",
"value": "AZS Olsztyn (volleyball)"
},
{
"operation": "back-button-clicked",
"time": "207.936",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "208.002",
"value": "2007–08 Polish Volleyball League"
},
{
"operation": "hyperlink",
"time": "209.605",
"value": "CEV Champions League"
},
{
"operation": "Now on",
"time": "210.859",
"value": "CEV Champions League"
},
{
"operation": "back-button-clicked",
"time": "214.571",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "214.645",
"value": "2007–08 Polish Volleyball League"
},
{
"operation": "finish",
"time": "230.785",
"value": "finish"
}
]
}
|
exfever_3702
|
The Economic Times is an Indian English-language business-focused daily newspaper published simultaneously in 15 cities, including a city and a union territory of India whose city proper population was over 11 million according to the 2011 census.
|
{
"evidence": [
"The Economic Times",
"Delhi"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Economic Times is an Indian English-language business-focused daily newspaper published simultaneously in 15 cities, including a city and a union territory of India whose city proper population was over 11 million according to the 2011 census.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Economic Times",
"Delhi"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6201,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
82225
|
Held at Toronto's Maple Leaf Gardens in order to maximize profits, the 1944 Memorial Cup was the 26th junior ice hockey championship of its kind.
|
{
"evidence": [
"1944 Memorial Cup_sentence_0",
"1944 Memorial Cup_sentence_1",
"1944 Memorial Cup_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nHeld at Toronto's Maple Leaf Gardens in order to maximize profits, the 1944 Memorial Cup was the 26th junior ice hockey championship of its kind.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"1944 Memorial Cup_sentence_0",
"1944 Memorial Cup_sentence_1",
"1944 Memorial Cup_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6202,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.845",
"value": "?search="
},
{
"operation": "search",
"time": "28.596",
"value": "1944 Memorial Cup"
},
{
"operation": "Now on",
"time": "29.692",
"value": "1944 Memorial Cup"
},
{
"operation": "Highlighting",
"time": "32.596",
"value": "1944 Memorial Cup_sentence_0"
},
{
"operation": "Highlighting",
"time": "34.162",
"value": "1944 Memorial Cup_sentence_1"
},
{
"operation": "Highlighting",
"time": "58.156",
"value": "1944 Memorial Cup_sentence_2"
},
{
"operation": "finish",
"time": "78.97",
"value": "finish"
}
]
}
|
34477
|
Jane Misme believed Red Cross nurses who took the job to find a husband were detrimental to the cause of women's suffrage.
|
{
"evidence": [
"Jane Misme_sentence_40",
"Jane Misme_sentence_41"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJane Misme believed Red Cross nurses who took the job to find a husband were detrimental to the cause of women's suffrage.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Jane Misme_sentence_40",
"Jane Misme_sentence_41"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6203,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.888",
"value": "?search="
},
{
"operation": "search",
"time": "12.436",
"value": "Jane Misme"
},
{
"operation": "Now on",
"time": "13.375",
"value": "Jane Misme"
},
{
"operation": "Highlighting",
"time": "216.816",
"value": "Jane Misme_sentence_40"
},
{
"operation": "Highlighting",
"time": "217.676",
"value": "Jane Misme_sentence_41"
},
{
"operation": "finish",
"time": "279.087",
"value": "finish"
}
]
}
|
exfever_1214
|
Sharon Marie Tate Polanski was an American model who appeared in the American mystery horror film directed by J. Lee Thompson.
|
{
"evidence": [
"Sharon Tate",
"Eye of the Devil"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nSharon Marie Tate Polanski was an American model who appeared in the American mystery horror film directed by J. Lee Thompson.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Sharon Tate",
"Eye of the Devil"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6204,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
71873
|
Born in 1997, Ferdinand Zvonimir von Habsburg, an Austrian motor racing driver and heir apparent to the headship of the House of Habsburg as eldest son of Karl von Habsburg, was the youngest individual in 2017 to be conferred the knightly Order of the Golden Fleece.
|
{
"evidence": [
"Ferdinand Zvonimir von Habsburg_sentence_0"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBorn in 1997, Ferdinand Zvonimir von Habsburg, an Austrian motor racing driver and heir apparent to the headship of the House of Habsburg as eldest son of Karl von Habsburg, was the youngest individual in 2017 to be conferred the knightly Order of the Golden Fleece.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Ferdinand Zvonimir von Habsburg_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6205,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.765",
"value": "?search="
},
{
"operation": "search",
"time": "38.079",
"value": "Ferdinand Zvonimir von Habsburg"
},
{
"operation": "Now on",
"time": "39.769",
"value": "Ferdinand Zvonimir von Habsburg"
},
{
"operation": "Highlighting",
"time": "47.612",
"value": "Ferdinand Zvonimir von Habsburg_sentence_0"
},
{
"operation": "hyperlink",
"time": "304.276",
"value": "Karl von Habsburg"
},
{
"operation": "Now on",
"time": "305.422",
"value": "Karl von Habsburg"
},
{
"operation": "back-button-clicked",
"time": "320.899",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "321.714",
"value": "Ferdinand Zvonimir von Habsburg"
},
{
"operation": "search",
"time": "329.976",
"value": "Contains...knightly Order of the Golden Fleece"
},
{
"operation": "Now on",
"time": "332.114",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "362.219",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "363.019",
"value": "Ferdinand Zvonimir von Habsburg"
},
{
"operation": "finish",
"time": "389.378",
"value": "finish"
}
]
}
|
44921
|
Nijō Mitsuhira, son of Nijō Yasumichi, held regent positions kampaku (a sort of chief advisor for the Emperor, but was the title of both first secretary and regent who assists an adult Emperor) from 1653 to 1666 and sesshō from 1663 to 1664.
|
{
"evidence": [
"Nijō Mitsuhira_sentence_2"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nNijō Mitsuhira, son of Nijō Yasumichi, held regent positions kampaku (a sort of chief advisor for the Emperor, but was the title of both first secretary and regent who assists an adult Emperor) from 1653 to 1666 and sesshō from 1663 to 1664.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Nijō Mitsuhira_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6206,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.405",
"value": "?search="
},
{
"operation": "search",
"time": "14.077",
"value": "Nijō Mitsuhira"
},
{
"operation": "Now on",
"time": "14.78",
"value": "Nijō Mitsuhira"
},
{
"operation": "Highlighting",
"time": "73.404",
"value": "Nijō Mitsuhira_sentence_1"
},
{
"operation": "Highlighting",
"time": "80.47",
"value": "Nijō Mitsuhira_sentence_2"
},
{
"operation": "hyperlink",
"time": "90.716",
"value": "kampaku"
},
{
"operation": "Now on",
"time": "92.33",
"value": "Sesshō and Kampaku"
},
{
"operation": "Highlighting",
"time": "120.894",
"value": "Sesshō and Kampaku_sentence_3"
},
{
"operation": "Highlighting",
"time": "120.895",
"value": "Sesshō and Kampaku_title"
},
{
"operation": "back-button-clicked",
"time": "141.52",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "141.573",
"value": "Nijō Mitsuhira"
},
{
"operation": "finish",
"time": "180.008",
"value": "finish"
}
]
}
|
85793
|
The Contribution of the Aviation Industry to the UK Economy was published in 1999, The Economic Contribution of the Aviation Industry in the UK was published in October 2006, and in December 2006 the government published the Air Transport White Paper Progress Report 2006 to report on progress made in "… delivering a sustainable future for aviation."
|
{
"evidence": [
"Air transport in the United Kingdom_sentence_57"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Contribution of the Aviation Industry to the UK Economy was published in 1999, The Economic Contribution of the Aviation Industry in the UK was published in October 2006, and in December 2006 the government published the Air Transport White Paper Progress Report 2006 to report on progress made in \"… delivering a sustainable future for aviation.\"\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Air transport in the United Kingdom_sentence_57"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6207,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.323",
"value": "?search="
},
{
"operation": "search",
"time": "421.337",
"value": "Contribution of the Aviation Industry to the UK Economy"
},
{
"operation": "Now on",
"time": "423.933",
"value": "undefined"
},
{
"operation": "search",
"time": "489.513",
"value": "The Economic Contribution of the Aviation Industry in the UK"
},
{
"operation": "Now on",
"time": "491.003",
"value": "undefined"
},
{
"operation": "search",
"time": "546.572",
"value": "Air Transport White Paper Progress Report 2006"
},
{
"operation": "Now on",
"time": "548.59",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "563.333",
"value": "Air transport in the United Kingdom"
},
{
"operation": "Now on",
"time": "564.175",
"value": "Air transport in the United Kingdom"
},
{
"operation": "Page search",
"time": "586.908",
"value": "2006"
},
{
"operation": "Highlighting",
"time": "637.184",
"value": "Air transport in the United Kingdom_sentence_57"
},
{
"operation": "Page search",
"time": "739.213",
"value": "economic"
},
{
"operation": "search",
"time": "818.438",
"value": "The Economic Contribution of the Aviation Industry in the UK"
},
{
"operation": "Now on",
"time": "819.686",
"value": "undefined"
},
{
"operation": "finish",
"time": "1012.295",
"value": "finish"
}
]
}
|
10312
|
Koopmans, born in Amsterdam, organised his first Alternative Elfstedentocht in 1975 in Finland due to the winters in the Netherlands being too warm to skate 200 km in open air on ice; then later, the Alternative Elfstedentochten were organised in various countries.
|
{
"evidence": [
"Aart Koopmans_sentence_2",
"Aart Koopmans_sentence_3"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nKoopmans, born in Amsterdam, organised his first Alternative Elfstedentocht in 1975 in Finland due to the winters in the Netherlands being too warm to skate 200 km in open air on ice; then later, the Alternative Elfstedentochten were organised in various countries.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Aart Koopmans_sentence_2",
"Aart Koopmans_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6208,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.627",
"value": "?search="
},
{
"operation": "search",
"time": "11.492",
"value": "Contains...Alternative Elfstedentocht"
},
{
"operation": "Now on",
"time": "12.219",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "17.818",
"value": "Elfstedentocht"
},
{
"operation": "Now on",
"time": "19.276",
"value": "Elfstedentocht"
},
{
"operation": "back-button-clicked",
"time": "22.915",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "23.483",
"value": "undefined"
},
{
"operation": "search",
"time": "28.877",
"value": "Koopmans"
},
{
"operation": "Now on",
"time": "31.034",
"value": "Koopmans"
},
{
"operation": "search",
"time": "42.144",
"value": "Contains...Alternative Elfstedentocht"
},
{
"operation": "Now on",
"time": "43.053",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "45.747",
"value": "Alternative Elfstedentocht Elfstedentocht_section_7"
},
{
"operation": "Now on",
"time": "46.212",
"value": "Elfstedentocht#Alternative Elfstedentocht Elfstedentocht section 7"
},
{
"operation": "Page search",
"time": "75.548",
"value": "koopmans"
},
{
"operation": "search",
"time": "101.216",
"value": "Koopmans"
},
{
"operation": "Now on",
"time": "102.543",
"value": "Koopmans"
},
{
"operation": "hyperlink",
"time": "112.918",
"value": "Leo Koopmans"
},
{
"operation": "Now on",
"time": "113.608",
"value": "Leo Koopmans"
},
{
"operation": "Page search",
"time": "119.21",
"value": "Alternative Elfstedentocht"
},
{
"operation": "back-button-clicked",
"time": "124.503",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "124.594",
"value": "Koopmans"
},
{
"operation": "hyperlink",
"time": "132.232",
"value": "Koopman"
},
{
"operation": "Now on",
"time": "132.867",
"value": "Koopman"
},
{
"operation": "Highlighting",
"time": "139.755",
"value": "Koopman_sentence_0"
},
{
"operation": "Highlighting",
"time": "139.757",
"value": "Koopman_title"
},
{
"operation": "Highlighting deleted",
"time": "146.417",
"value": "Koopman_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "146.418",
"value": "Koopman_title"
},
{
"operation": "search",
"time": "178.576",
"value": "Contains...Alternative Elfstedentocht"
},
{
"operation": "Now on",
"time": "179.835",
"value": "undefined"
},
{
"operation": "Page search",
"time": "186.694",
"value": "koopmans"
},
{
"operation": "hyperlink",
"time": "190.448",
"value": "Elfstedentocht Speed skating_section_3"
},
{
"operation": "Now on",
"time": "191.55",
"value": "Speed skating#Elfstedentocht Speed skating section 3"
},
{
"operation": "Page search",
"time": "205.231",
"value": "koopmans"
},
{
"operation": "search",
"time": "227.505",
"value": "Contains...Alternative Elfstedentocht"
},
{
"operation": "Now on",
"time": "228.276",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "237.119",
"value": "Aart Koopmans"
},
{
"operation": "Now on",
"time": "238.704",
"value": "Aart Koopmans"
},
{
"operation": "Page search",
"time": "246.745",
"value": "Alternative Elfstedentocht"
},
{
"operation": "Highlighting",
"time": "251.042",
"value": "Aart Koopmans_sentence_2"
},
{
"operation": "Highlighting",
"time": "258.699",
"value": "Aart Koopmans_sentence_3"
},
{
"operation": "finish",
"time": "279.723",
"value": "finish"
}
]
}
|
exfever_3379
|
The second season of Psych originally aired in the United States with the appearance of an American actor who is notable for his supporting roles as Sean Potter on the CBS television series Judging Amy.
|
{
"evidence": [
"Psych (season 2)",
"Timothy Omundson"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe second season of Psych originally aired in the United States with the appearance of an American actor who is notable for his supporting roles as Sean Potter on the CBS television series Judging Amy.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Psych (season 2)",
"Timothy Omundson"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6209,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7741
|
The invasion of North Africa in November 1942 was intended to draw Axis forces away from the Eastern Front.
|
{
"evidence": [
"Dwight D. Eisenhower",
"Invasion of Normandy"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe invasion of North Africa in November 1942 was intended to draw Axis forces away from the Eastern Front.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Dwight D. Eisenhower",
"Invasion of Normandy"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6210,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2409
|
Inferno is a 2016 film starring an Australian actor and filmmaker who made his breakthrough with a leading role in a 1992 sports comedy-drama film.
|
{
"evidence": [
"Inferno (2016 film)",
"Tom Hanks",
"A League of Their Own"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nInferno is a 2016 film starring an Australian actor and filmmaker who made his breakthrough with a leading role in a 1992 sports comedy-drama film.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Inferno (2016 film)",
"Tom Hanks",
"A League of Their Own"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6211,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
57222
|
Gary Hunt of Great Britain won the high diving event at the 2015 World Aquatics Championships – Men held on August 3 and 5.
|
{
"evidence": [
"Gary Hunt_sentence_0"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nGary Hunt of Great Britain won the high diving event at the 2015 World Aquatics Championships – Men held on August 3 and 5.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Gary Hunt_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6212,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.108",
"value": "?search="
},
{
"operation": "search",
"time": "15.63",
"value": "Gary Hunt"
},
{
"operation": "Now on",
"time": "16.47",
"value": "Gary Hunt"
},
{
"operation": "Page search",
"time": "34.871",
"value": "2015"
},
{
"operation": "Highlighting",
"time": "57.103",
"value": "Gary Hunt_sentence_0"
},
{
"operation": "Page search",
"time": "94.186",
"value": "great britain"
},
{
"operation": "Page search",
"time": "114.491",
"value": "born"
},
{
"operation": "finish",
"time": "129.368",
"value": "finish"
}
]
}
|
78347
|
Riblja Čorba leader, Bora Đorđević, not only started performing and composing as a teenager; but also sang and played rhythm guitar in his early bands including Poslednji Ostatak Romantičnog Sveta.
|
{
"evidence": [
"Bora Đorđević_sentence_2",
"Bora Đorđević_sentence_5",
"Bora Đorđević_sentence_6",
"Bora Đorđević_sentence_8",
"Bora Đorđević_sentence_11",
"Bora Đorđević_sentence_12"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nRiblja Čorba leader, Bora Đorđević, not only started performing and composing as a teenager; but also sang and played rhythm guitar in his early bands including Poslednji Ostatak Romantičnog Sveta.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Bora Đorđević_sentence_2",
"Bora Đorđević_sentence_5",
"Bora Đorđević_sentence_6",
"Bora Đorđević_sentence_8",
"Bora Đorđević_sentence_11",
"Bora Đorđević_sentence_12"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6213,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.88",
"value": "?search="
},
{
"operation": "search",
"time": "446.829",
"value": "Bora Đorđević"
},
{
"operation": "Now on",
"time": "448.285",
"value": "Bora Đorđević"
},
{
"operation": "Highlighting",
"time": "458.16",
"value": "Bora Đorđević_sentence_2"
},
{
"operation": "Highlighting",
"time": "556.209",
"value": "Bora Đorđević_sentence_5"
},
{
"operation": "Highlighting",
"time": "570.706",
"value": "Bora Đorđević_sentence_8"
},
{
"operation": "Highlighting",
"time": "600.116",
"value": "Bora Đorđević_sentence_6"
},
{
"operation": "Highlighting deleted",
"time": "601.087",
"value": "Bora Đorđević_sentence_6"
},
{
"operation": "Highlighting deleted",
"time": "602.391",
"value": "Bora Đorđević_sentence_8"
},
{
"operation": "Highlighting",
"time": "604.14",
"value": "Bora Đorđević_sentence_8"
},
{
"operation": "search",
"time": "651.425",
"value": "Contains...Poslednji Ostatak Romantičnog Sveta"
},
{
"operation": "Now on",
"time": "652.613",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "656.606",
"value": "Bora Đorđević"
},
{
"operation": "Now on",
"time": "656.91",
"value": "Bora Đorđević"
},
{
"operation": "Highlighting",
"time": "668.123",
"value": "Bora Đorđević_sentence_11"
},
{
"operation": "Highlighting",
"time": "671.851",
"value": "Bora Đorđević_sentence_12"
},
{
"operation": "Highlighting deleted",
"time": "687.581",
"value": "Bora Đorđević_sentence_12"
},
{
"operation": "Highlighting deleted",
"time": "688.101",
"value": "Bora Đorđević_sentence_11"
},
{
"operation": "Highlighting deleted",
"time": "688.59",
"value": "Bora Đorđević_sentence_8"
},
{
"operation": "Highlighting",
"time": "689.213",
"value": "Bora Đorđević_sentence_6"
},
{
"operation": "Highlighting",
"time": "692.01",
"value": "Bora Đorđević_sentence_8"
},
{
"operation": "Highlighting",
"time": "697.501",
"value": "Bora Đorđević_sentence_11"
},
{
"operation": "Highlighting",
"time": "698.137",
"value": "Bora Đorđević_sentence_12"
},
{
"operation": "finish",
"time": "729.803",
"value": "finish"
}
]
}
|
exfever_8486
|
Reba Nell McEntire is an American country music singer who was interested during her childhood in a rodeo event in which a horse and rider attempted to run a cloverleaf pattern around preset barrels in the fastest time.
|
{
"evidence": [
"Reba McEntire",
"Barrel racing"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nReba Nell McEntire is an American country music singer who was interested during her childhood in a rodeo event in which a horse and rider attempted to run a cloverleaf pattern around preset barrels in the fastest time.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Reba McEntire",
"Barrel racing"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6214,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2593
|
Paul von Hindenburg was a German Field Marshal and statesman who was appointed as Chancellor of Germany by an Austrian-born German politician who was the dictator of Germany.
|
{
"evidence": [
"Paul von Hindenburg",
"Adolf Hitler"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nPaul von Hindenburg was a German Field Marshal and statesman who was appointed as Chancellor of Germany by an Austrian-born German politician who was the dictator of Germany.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Paul von Hindenburg",
"Adolf Hitler"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6215,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
14071
|
Some Orphic poems were ascribed to Brontinus, and among which was a poem On Nature (Physika).
|
{
"evidence": [
"Brontinus_sentence_3",
"Brontinus_sentence_4"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nSome Orphic poems were ascribed to Brontinus, and among which was a poem On Nature (Physika).\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Brontinus_sentence_3",
"Brontinus_sentence_4"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6216,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.941",
"value": "?search="
},
{
"operation": "search",
"time": "14.938",
"value": "Brontinus"
},
{
"operation": "Now on",
"time": "16.385",
"value": "Brontinus"
},
{
"operation": "Highlighting",
"time": "23.086",
"value": "Brontinus_sentence_3"
},
{
"operation": "Highlighting",
"time": "29.723",
"value": "Brontinus_sentence_4"
},
{
"operation": "finish",
"time": "44.396",
"value": "finish"
}
]
}
|
49886
|
The Eucrostes disparata moth was first described in 1861 by Francis Walker, an English entomologist and prolific author.
|
{
"evidence": [
"Eucrostes disparata_sentence_0",
"Francis Walker (entomologist)_sentence_0",
"Francis Walker (entomologist)_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Eucrostes disparata moth was first described in 1861 by Francis Walker, an English entomologist and prolific author.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Eucrostes disparata_sentence_0",
"Francis Walker (entomologist)_sentence_0",
"Francis Walker (entomologist)_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6217,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.095",
"value": "?search="
},
{
"operation": "search",
"time": "31.415",
"value": "Eucrostes disparata"
},
{
"operation": "Now on",
"time": "31.938",
"value": "Eucrostes disparata"
},
{
"operation": "Highlighting",
"time": "35.23",
"value": "Eucrostes disparata_sentence_0"
},
{
"operation": "hyperlink",
"time": "44.022",
"value": "Francis Walker"
},
{
"operation": "Now on",
"time": "44.766",
"value": "Francis Walker (entomologist)"
},
{
"operation": "Highlighting",
"time": "49.447",
"value": "Francis Walker (entomologist)_sentence_0"
},
{
"operation": "Highlighting",
"time": "56.094",
"value": "Francis Walker (entomologist)_sentence_2"
},
{
"operation": "finish",
"time": "152.654",
"value": "finish"
}
]
}
|
88536
|
Nagor Amir Noor Mohamed is an international association baseball referee who officiated numerous matches in the AFC Champions League and the AFC Cup, including the 2012 AFC President's Cup, 2014 AFC Challenge Cup and 2014 WAFF Championship.
|
{
"evidence": [
"Nagor Amir Noor Mohamed_sentence_0"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nNagor Amir Noor Mohamed is an international association baseball referee who officiated numerous matches in the AFC Champions League and the AFC Cup, including the 2012 AFC President's Cup, 2014 AFC Challenge Cup and 2014 WAFF Championship.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Nagor Amir Noor Mohamed_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6218,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.693",
"value": "?search="
},
{
"operation": "search",
"time": "8.79",
"value": "Nagor Amir Noor Mohamed"
},
{
"operation": "Now on",
"time": "9.491",
"value": "Nagor Amir Noor Mohamed"
},
{
"operation": "Highlighting",
"time": "153.13",
"value": "Nagor Amir Noor Mohamed_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "153.228",
"value": "Nagor Amir Noor Mohamed_sentence_0"
},
{
"operation": "Highlighting",
"time": "153.963",
"value": "Nagor Amir Noor Mohamed_sentence_0"
},
{
"operation": "hyperlink",
"time": "162.519",
"value": "association football"
},
{
"operation": "Now on",
"time": "164.162",
"value": "Association football"
},
{
"operation": "Now on",
"time": "169.228",
"value": "Association football"
},
{
"operation": "finish",
"time": "175.364",
"value": "finish"
}
]
}
|
17732
|
Faris Shah Rosli scored more goals for Harimau Muda than for Kelantan and Melaka United which competes in the Chinese Super League.
|
{
"evidence": [
"Melaka United_sentence_1"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nFaris Shah Rosli scored more goals for Harimau Muda than for Kelantan and Melaka United which competes in the Chinese Super League.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Melaka United_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6219,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.663",
"value": "?search="
},
{
"operation": "search",
"time": "22.294",
"value": "Faris Shah Rosli"
},
{
"operation": "Now on",
"time": "24.649",
"value": "Faris Shah Rosli"
},
{
"operation": "Highlighting",
"time": "40.35",
"value": "Faris Shah Rosli_cell_0_4_0"
},
{
"operation": "Highlighting",
"time": "40.351",
"value": "Faris Shah Rosli_title"
},
{
"operation": "Highlighting",
"time": "40.351",
"value": "Faris Shah Rosli_section_6"
},
{
"operation": "Highlighting",
"time": "40.352",
"value": "Faris Shah Rosli_section_7"
},
{
"operation": "Highlighting",
"time": "40.432",
"value": "Faris Shah Rosli_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "47.715",
"value": "Faris Shah Rosli_header_cell_0_5_3"
},
{
"operation": "Highlighting",
"time": "48.918",
"value": "Faris Shah Rosli_header_cell_0_1_2"
},
{
"operation": "Highlighting",
"time": "53.787",
"value": "Faris Shah Rosli_header_cell_0_5_0"
},
{
"operation": "Highlighting",
"time": "60.033",
"value": "Faris Shah Rosli_header_cell_0_5_11"
},
{
"operation": "Highlighting",
"time": "61.183",
"value": "Faris Shah Rosli_header_cell_0_1_10"
},
{
"operation": "Highlighting deleted",
"time": "62.401",
"value": "Faris Shah Rosli_header_cell_0_5_3"
},
{
"operation": "Highlighting deleted",
"time": "63.087",
"value": "Faris Shah Rosli_header_cell_0_1_2"
},
{
"operation": "Highlighting",
"time": "77.307",
"value": "Faris Shah Rosli_cell_0_6_0"
},
{
"operation": "Highlighting",
"time": "80.345",
"value": "Faris Shah Rosli_header_cell_0_9_11"
},
{
"operation": "Highlighting",
"time": "84.157",
"value": "Faris Shah Rosli_header_cell_0_9_0"
},
{
"operation": "Highlighting",
"time": "87.459",
"value": "Faris Shah Rosli_cell_0_10_0"
},
{
"operation": "Highlighting",
"time": "89.095",
"value": "Faris Shah Rosli_header_cell_0_12_0"
},
{
"operation": "Highlighting",
"time": "90.221",
"value": "Faris Shah Rosli_header_cell_0_12_11"
},
{
"operation": "hyperlink",
"time": "96.993",
"value": "Melaka United"
},
{
"operation": "Now on",
"time": "100.01",
"value": "Melaka United"
},
{
"operation": "Highlighting",
"time": "101.547",
"value": "Melaka United_sentence_1"
},
{
"operation": "back-button-clicked",
"time": "104.694",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "104.835",
"value": "Faris Shah Rosli"
},
{
"operation": "Highlighting deleted",
"time": "106.607",
"value": "Faris Shah Rosli_cell_0_10_0"
},
{
"operation": "Highlighting deleted",
"time": "107.06",
"value": "Faris Shah Rosli_cell_0_6_0"
},
{
"operation": "Highlighting deleted",
"time": "108.095",
"value": "Faris Shah Rosli_cell_0_4_0"
},
{
"operation": "Highlighting deleted",
"time": "108.165",
"value": "Faris Shah Rosli_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "108.743",
"value": "Faris Shah Rosli_header_cell_0_5_0"
},
{
"operation": "Highlighting deleted",
"time": "109.291",
"value": "Faris Shah Rosli_header_cell_0_9_0"
},
{
"operation": "Highlighting deleted",
"time": "110.07",
"value": "Faris Shah Rosli_header_cell_0_12_0"
},
{
"operation": "Highlighting deleted",
"time": "111.013",
"value": "Faris Shah Rosli_header_cell_0_5_11"
},
{
"operation": "Highlighting deleted",
"time": "111.565",
"value": "Faris Shah Rosli_header_cell_0_9_11"
},
{
"operation": "Highlighting deleted",
"time": "112.191",
"value": "Faris Shah Rosli_header_cell_0_12_11"
},
{
"operation": "Highlighting deleted",
"time": "114.212",
"value": "Faris Shah Rosli_header_cell_0_1_10"
},
{
"operation": "Highlighting deleted",
"time": "114.214",
"value": "Faris Shah Rosli_title"
},
{
"operation": "Highlighting deleted",
"time": "114.216",
"value": "Faris Shah Rosli_section_6"
},
{
"operation": "Highlighting deleted",
"time": "114.217",
"value": "Faris Shah Rosli_section_7"
},
{
"operation": "finish",
"time": "130.16",
"value": "finish"
}
]
}
|
exfever_6816
|
Roman Holiday is the first romantic comedy film where Audrey Hepburn appeared.
|
{
"evidence": [
"Audrey Hepburn",
"Ixelles"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nRoman Holiday is the first romantic comedy film where Audrey Hepburn appeared.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Audrey Hepburn",
"Ixelles"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6220,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2859
|
Golf is a club-and-ball sport that was first recorded in the Scottish late Middle Ages, and the modern game of golf was first developed and established in the country.
|
{
"evidence": [
"Golf",
"Golf in Scotland"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nGolf is a club-and-ball sport that was first recorded in the Scottish late Middle Ages, and the modern game of golf was first developed and established in the country.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Golf",
"Golf in Scotland"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6221,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8027
|
30 Rock is a Canadian satirical television series that originally aired on an American English-language commercial broadcast television and radio network that is owned by Comcast through NBCUniversal.
|
{
"evidence": [
"30 Rock",
"NBC"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\n30 Rock is a Canadian satirical television series that originally aired on an American English-language commercial broadcast television and radio network that is owned by Comcast through NBCUniversal.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"30 Rock",
"NBC"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6222,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4303
|
Zach_Galifianakis appeared in Comedy Central Presents special and presented his television show on VH1 during the spring of 2005.
|
{
"evidence": [
"Zach Galifianakis",
"Late World with Zach"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nZach_Galifianakis appeared in Comedy Central Presents special and presented his television show on VH1 during the spring of 2005.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Zach Galifianakis",
"Late World with Zach"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6223,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_31
|
The Lion King is an Australian animated musical drama film directed by an American filmmaker known for directing Stuart Little, Stuart Little 2, The Haunted Mansion, The Forbidden Kingdom, and Mr. Peabody & Sherman.
|
{
"evidence": [
"The Lion King",
"Rob Minkoff"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Lion King is an Australian animated musical drama film directed by an American filmmaker known for directing Stuart Little, Stuart Little 2, The Haunted Mansion, The Forbidden Kingdom, and Mr. Peabody & Sherman.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Lion King",
"Rob Minkoff"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6224,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7752
|
Dracula, which was published in 1897, is a comic written as a series of documents.
|
{
"evidence": [
"Dracula",
"Epistolary novel"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nDracula, which was published in 1897, is a comic written as a series of documents.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Dracula",
"Epistolary novel"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6225,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8779
|
The Jackson family is an American family of musicians that are sometimes called the "First Family of Soul", which was first bestowed on a soul group made up of five of Betty and Clarence Burke Sr.'s six children, who are all citizens of the United States of America.
|
{
"evidence": [
"Jackson family",
"Five Stairsteps",
"Americans"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Jackson family is an American family of musicians that are sometimes called the \"First Family of Soul\", which was first bestowed on a soul group made up of five of Betty and Clarence Burke Sr.'s six children, who are all citizens of the United States of America.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Jackson family",
"Five Stairsteps",
"Americans"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6226,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3965
|
Pineapple Express is a 2008 American stoner comedy film directed by an American filmmaker who also directed a 2011 American dark comedy novel.
|
{
"evidence": [
"Pineapple Express (film)",
"David Gordon Green",
"The Sitter"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nPineapple Express is a 2008 American stoner comedy film directed by an American filmmaker who also directed a 2011 American dark comedy novel.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Pineapple Express (film)",
"David Gordon Green",
"The Sitter"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6227,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8260
|
Thirteen is a 2003 African teen drama film starring an American actress who is the recipient of a Critics' Choice Television Award as well as three Primetime Emmy Award nominations and three Golden Globe Award nominations for her work in film and television.
|
{
"evidence": [
"Thirteen (2003 film)",
"Evan Rachel Wood"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThirteen is a 2003 African teen drama film starring an American actress who is the recipient of a Critics' Choice Television Award as well as three Primetime Emmy Award nominations and three Golden Globe Award nominations for her work in film and television.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Thirteen (2003 film)",
"Evan Rachel Wood"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6228,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
73394
|
Steven Nitah is a Canadian politician who first ran for a seat in the 1999 Northwest Territories general election, he ran in that election with the backing of former Premier Don Morin and ran for re-election in the 2003 Northwest Territories general election.
|
{
"evidence": [
"Steven Nitah_sentence_0",
"Steven Nitah_sentence_1",
"Steven Nitah_sentence_2",
"Steven Nitah_sentence_3"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nSteven Nitah is a Canadian politician who first ran for a seat in the 1999 Northwest Territories general election, he ran in that election with the backing of former Premier Don Morin and ran for re-election in the 2003 Northwest Territories general election.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Steven Nitah_sentence_0",
"Steven Nitah_sentence_1",
"Steven Nitah_sentence_2",
"Steven Nitah_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6229,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.168",
"value": "?search="
},
{
"operation": "search",
"time": "10.329",
"value": "Steven Nitah"
},
{
"operation": "Now on",
"time": "11.183",
"value": "Steven Nitah"
},
{
"operation": "Highlighting",
"time": "14.695",
"value": "Steven Nitah_sentence_0"
},
{
"operation": "Highlighting",
"time": "17.497",
"value": "Steven Nitah_sentence_1"
},
{
"operation": "Highlighting",
"time": "21.271",
"value": "Steven Nitah_sentence_2"
},
{
"operation": "Highlighting",
"time": "26.243",
"value": "Steven Nitah_sentence_3"
},
{
"operation": "finish",
"time": "52.277",
"value": "finish"
}
]
}
|
exfever_1539
|
Linkin Park is an English rock band from a city in the Santa Monica Mountains region of Los Angeles County.
|
{
"evidence": [
"Linkin Park",
"Agoura Hills, California"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nLinkin Park is an English rock band from a city in the Santa Monica Mountains region of Los Angeles County.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Linkin Park",
"Agoura Hills, California"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6230,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
84645
|
Margaret Morgan Lawrence was self-educated.
|
{
"evidence": [
"Margaret Morgan Lawrence_sentence_9",
"Margaret Morgan Lawrence_sentence_10"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nMargaret Morgan Lawrence was self-educated.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Margaret Morgan Lawrence_sentence_9",
"Margaret Morgan Lawrence_sentence_10"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6231,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.026",
"value": "?search="
},
{
"operation": "search",
"time": "42.366",
"value": "Margaret Morgan Lawrence"
},
{
"operation": "Now on",
"time": "43.065",
"value": "Margaret Morgan Lawrence"
},
{
"operation": "Highlighting",
"time": "168.717",
"value": "Margaret Morgan Lawrence_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "171.277",
"value": "Margaret Morgan Lawrence_sentence_0"
},
{
"operation": "Page search",
"time": "178.261",
"value": "self"
},
{
"operation": "Page search",
"time": "178.365",
"value": "self"
},
{
"operation": "Highlighting",
"time": "188.013",
"value": "Margaret Morgan Lawrence_sentence_9"
},
{
"operation": "Highlighting",
"time": "191.221",
"value": "Margaret Morgan Lawrence_sentence_10"
},
{
"operation": "Page search",
"time": "219.877",
"value": "educated"
},
{
"operation": "finish",
"time": "234.254",
"value": "finish"
}
]
}
|
exfever_4713
|
Eric Trump is the third son of an American politician, a media personality, and a businessman.
|
{
"evidence": [
"Eric Trump",
"Donald Trump"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nEric Trump is the third son of an American politician, a media personality, and a businessman.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Eric Trump",
"Donald Trump"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6232,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
82741
|
Mode Media was founded by nine people in 2003 and was originally a website that offered fashion and beauty content and blog content.
|
{
"evidence": [
"Mode Media_sentence_5",
"Mode Media_sentence_6"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nMode Media was founded by nine people in 2003 and was originally a website that offered fashion and beauty content and blog content.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Mode Media_sentence_5",
"Mode Media_sentence_6"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6233,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.874",
"value": "?search="
},
{
"operation": "search",
"time": "243.71",
"value": "Mode Media"
},
{
"operation": "Now on",
"time": "244.83",
"value": "Mode Media"
},
{
"operation": "Highlighting",
"time": "262.292",
"value": "Mode Media_sentence_5"
},
{
"operation": "Highlighting",
"time": "273.462",
"value": "Mode Media_sentence_6"
},
{
"operation": "finish",
"time": "283.504",
"value": "finish"
}
]
}
|
exfever_7681
|
Coldplay is a British rock band that signed with a German–British record label founded in America that developed a reputation as a record label that originated in the African-American communities of New Orleans, Louisiana, with roots in blues and ragtime.
|
{
"evidence": [
"Coldplay",
"Parlophone",
"Jazz"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nColdplay is a British rock band that signed with a German–British record label founded in America that developed a reputation as a record label that originated in the African-American communities of New Orleans, Louisiana, with roots in blues and ragtime.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Coldplay",
"Parlophone",
"Jazz"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6234,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
2090
|
Doris Merrick, who was born in June 1919 to Joseph Merrick and Nellie Simpson, worked as a singer, a model, and has also appeared in a film in 1941.
|
{
"evidence": [
"Doris Merrick_sentence_1",
"Doris Merrick_sentence_2",
"Doris Merrick_sentence_4"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nDoris Merrick, who was born in June 1919 to Joseph Merrick and Nellie Simpson, worked as a singer, a model, and has also appeared in a film in 1941.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Doris Merrick_sentence_1",
"Doris Merrick_sentence_2",
"Doris Merrick_sentence_4"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6235,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "search",
"time": "11.388",
"value": "Doris Merrick"
},
{
"operation": "Highlighting",
"time": "18.402",
"value": "Doris Merrick_sentence_1"
},
{
"operation": "Highlighting",
"time": "36.483",
"value": "Doris Merrick_sentence_2"
},
{
"operation": "Highlighting",
"time": "54.857",
"value": "Doris Merrick_sentence_4"
},
{
"operation": "finish",
"time": "62.805",
"value": "finish"
}
]
}
|
exfever_4948
|
Dead Man Down is a 2013 American fantasy film starring an actor who appeared in an American musical film focusing on blues musician Ray Charles.
|
{
"evidence": [
"Dead Man Down",
"Terrence Howard",
"Ray (film)"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nDead Man Down is a 2013 American fantasy film starring an actor who appeared in an American musical film focusing on blues musician Ray Charles.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Dead Man Down",
"Terrence Howard",
"Ray (film)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6236,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
72723
|
A specie of mushroom has a cap and hymenium.
|
{
"evidence": [
"Hymenium_sentence_0",
"Hymenium_sentence_4"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nA specie of mushroom has a cap and hymenium.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Hymenium_sentence_0",
"Hymenium_sentence_4"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6237,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.02",
"value": "?search="
},
{
"operation": "search",
"time": "12.33",
"value": "Hymenium"
},
{
"operation": "Now on",
"time": "13.023",
"value": "Hymenium"
},
{
"operation": "Highlighting",
"time": "143.908",
"value": "Hymenium_sentence_0"
},
{
"operation": "Highlighting",
"time": "197.525",
"value": "Hymenium_sentence_4"
},
{
"operation": "finish",
"time": "203.545",
"value": "finish"
}
]
}
|
exfever_997
|
Adam Sandler is an American comedian whose comedic roles include a 1996 American sports comedy film.
|
{
"evidence": [
"Adam Sandler",
"Happy Gilmore"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nAdam Sandler is an American comedian whose comedic roles include a 1996 American sports comedy film.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Adam Sandler",
"Happy Gilmore"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6238,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
84199
|
Out of the seven games of Wrexham on the route to the final, four games were held in The Racecourse, Bangor, while the other three games were done in all different locations.
|
{
"evidence": [
"Wrexham A.F.C._sentence_0"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nOut of the seven games of Wrexham on the route to the final, four games were held in The Racecourse, Bangor, while the other three games were done in all different locations.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Wrexham A.F.C._sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6239,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.155",
"value": "?search="
},
{
"operation": "search",
"time": "43.576",
"value": "Wrexham"
},
{
"operation": "Now on",
"time": "45.422",
"value": "Wrexham"
},
{
"operation": "search",
"time": "74.271",
"value": "Contains...Out of the seven games of Wrexham"
},
{
"operation": "Now on",
"time": "77.563",
"value": "undefined"
},
{
"operation": "search",
"time": "95.052",
"value": "Contains...four games were held in The Racecourse, Bangor,"
},
{
"operation": "Now on",
"time": "97.175",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "106.866",
"value": "Wrexham A.F.C."
},
{
"operation": "Now on",
"time": "109.505",
"value": "Wrexham A.F.C."
},
{
"operation": "Highlighting",
"time": "207.445",
"value": "Wrexham A.F.C._sentence_0"
},
{
"operation": "finish",
"time": "219.868",
"value": "finish"
}
]
}
|
exfever_1419
|
The Black Dahlia is a 2006 film whose stars include a Canadian actress.
|
{
"evidence": [
"The Black Dahlia (film)",
"Mia Kirshner"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Black Dahlia is a 2006 film whose stars include a Canadian actress.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Black Dahlia (film)",
"Mia Kirshner"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6240,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7206
|
Theodore Roosevelt Jr. was an Australian politician, statesman, conservationist, naturalist, historian, and writer who emerged as a leader of the Republican Party and became a driving force for a period of widespread social activism and political reform across the United States of America.
|
{
"evidence": [
"Theodore Roosevelt",
"Progressive Era"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nTheodore Roosevelt Jr. was an Australian politician, statesman, conservationist, naturalist, historian, and writer who emerged as a leader of the Republican Party and became a driving force for a period of widespread social activism and political reform across the United States of America.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Theodore Roosevelt",
"Progressive Era"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6241,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2199
|
The Bold and the Beautiful is an American television soap opera created by William J. Bell and Lee Phillip Bell for CBS as Bell’s other soap operas companion series which differ to a degree from a television program derived from already existing works that focus on more details and different aspects from the original work.
|
{
"evidence": [
"The Bold and the Beautiful",
"Sister show",
"Spin-off (media)"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Bold and the Beautiful is an American television soap opera created by William J. Bell and Lee Phillip Bell for CBS as Bell’s other soap operas companion series which differ to a degree from a television program derived from already existing works that focus on more details and different aspects from the original work.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Bold and the Beautiful",
"Sister show",
"Spin-off (media)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6242,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8414
|
Jean-Claude Van Damme became a popular action film star and followed up with a 1990 film directed by Sheldon Lettich.
|
{
"evidence": [
"Jean-Claude Van Damme",
"Lionheart (1990 film)"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJean-Claude Van Damme became a popular action film star and followed up with a 1990 film directed by Sheldon Lettich.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Jean-Claude Van Damme",
"Lionheart (1990 film)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6243,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6016
|
Joan Crawford's first movie was Lady of the Night.
|
{
"evidence": [
"Joan Crawford",
"Mildred Pierce (film)"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJoan Crawford's first movie was Lady of the Night.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Joan Crawford",
"Mildred Pierce (film)"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6244,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4602
|
John Fante was best known for his novel about the life of a struggling writer in the largest city in California.
|
{
"evidence": [
"John Fante",
"Los Angeles"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJohn Fante was best known for his novel about the life of a struggling writer in the largest city in California.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"John Fante",
"Los Angeles"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6245,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8675
|
Taipei City rests on the Taipei Basin, which is bounded by the relatively narrow valleys of the Tainan and a river in northern Taiwan that flows through New Taipei and the capital Taipei for 82 kilometers.
|
{
"evidence": [
"Taipei",
"Xindian River"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nTaipei City rests on the Taipei Basin, which is bounded by the relatively narrow valleys of the Tainan and a river in northern Taiwan that flows through New Taipei and the capital Taipei for 82 kilometers.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Taipei",
"Xindian River"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6246,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
46694
|
Alfonso, born on November 11, 1986, beat Newfel Ouatah but lost to Russian Evgeny Romanov at the World Junior Championships 2004; then was also defeated nationally by compatriot superstar Odlanier Solis twice in 2004.
|
{
"evidence": [
"Robert Alfonso_sentence_1",
"Robert Alfonso_sentence_3",
"Robert Alfonso_sentence_4"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nAlfonso, born on November 11, 1986, beat Newfel Ouatah but lost to Russian Evgeny Romanov at the World Junior Championships 2004; then was also defeated nationally by compatriot superstar Odlanier Solis twice in 2004.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Robert Alfonso_sentence_1",
"Robert Alfonso_sentence_3",
"Robert Alfonso_sentence_4"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6247,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.764",
"value": "?search="
},
{
"operation": "search",
"time": "20.306",
"value": "Alfonso, born on November 11, 1986"
},
{
"operation": "Now on",
"time": "24.922",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "32.133",
"value": "Robert Alfonso"
},
{
"operation": "Now on",
"time": "32.843",
"value": "Robert Alfonso"
},
{
"operation": "Highlighting",
"time": "37.423",
"value": "Robert Alfonso_sentence_1"
},
{
"operation": "Highlighting",
"time": "48.871",
"value": "Robert Alfonso_sentence_3"
},
{
"operation": "Highlighting",
"time": "56.259",
"value": "Robert Alfonso_sentence_4"
},
{
"operation": "finish",
"time": "64.384",
"value": "finish"
}
]
}
|
exfever_4013
|
Dakota Fanning is an American actress who starred in a 2006 American fantasy film based on the 1952 novel by E.B. Winchester.
|
{
"evidence": [
"Dakota Fanning",
"Charlotte's Web (2006 film)"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nDakota Fanning is an American actress who starred in a 2006 American fantasy film based on the 1952 novel by E.B. Winchester.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Dakota Fanning",
"Charlotte's Web (2006 film)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6248,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
90942
|
Frank Irving was a British aeronautical engineer, glider pilot which is a recreational activity and competitive air sport.
|
{
"evidence": [
"Frank Irving_sentence_2",
"Gliding_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nFrank Irving was a British aeronautical engineer, glider pilot which is a recreational activity and competitive air sport.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Frank Irving_sentence_2",
"Gliding_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6249,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.55",
"value": "?search="
},
{
"operation": "search",
"time": "15.269",
"value": "Frank Irving"
},
{
"operation": "Now on",
"time": "16.029",
"value": "Frank Irving"
},
{
"operation": "Highlighting",
"time": "22.35",
"value": "Frank Irving_sentence_2"
},
{
"operation": "hyperlink",
"time": "24.702",
"value": "glider pilot"
},
{
"operation": "Now on",
"time": "25.68",
"value": "Gliding"
},
{
"operation": "Highlighting",
"time": "28.873",
"value": "Gliding_sentence_2"
},
{
"operation": "finish",
"time": "44.289",
"value": "finish"
}
]
}
|
exfever_7592
|
Levellers is the first countercultures in the Western world.
|
{
"evidence": [
"Counterculture",
"Levellers",
"Political movement"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nLevellers is the first countercultures in the Western world.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Counterculture",
"Levellers",
"Political movement"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6250,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2706
|
Tobias Vincent Maguire is an American actor who has done films, including the film, set during the American Civil War, which follows a group of men who join the First Irregulars of the state in the Midwestern region of the United Kingdom.
|
{
"evidence": [
"Tobey Maguire",
"Ride with the Devil (film)",
"Missouri"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nTobias Vincent Maguire is an American actor who has done films, including the film, set during the American Civil War, which follows a group of men who join the First Irregulars of the state in the Midwestern region of the United Kingdom.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tobey Maguire",
"Ride with the Devil (film)",
"Missouri"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6251,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1846
|
Ji Chang-wook is a South Korean actor and singer who is notable for his lead roles in in television series such as a 2011 South Korean fusion historical/action television series aired on SBS from July 4 to October 10, 2011.
|
{
"evidence": [
"Ji Chang-wook",
"Warrior Baek Dong-soo"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJi Chang-wook is a South Korean actor and singer who is notable for his lead roles in in television series such as a 2011 South Korean fusion historical/action television series aired on SBS from July 4 to October 10, 2011.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Ji Chang-wook",
"Warrior Baek Dong-soo"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6252,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1430
|
Shonen Jump features serialized chapters from four manga series and articles on the Japanese language and culture, such as hand-drawn and computer-generated animations originating from Japan.
|
{
"evidence": [
"Shonen Jump (magazine)",
"Anime"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nShonen Jump features serialized chapters from four manga series and articles on the Japanese language and culture, such as hand-drawn and computer-generated animations originating from Japan.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Shonen Jump (magazine)",
"Anime"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6253,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2098
|
How to Train Your Dragon is a 2010 American computer-animated fantasy film with a film genre in which the protagonist is thrust into a series of events that typically involve violence and physical feats.
|
{
"evidence": [
"How to Train Your Dragon (film)",
"Action film"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nHow to Train Your Dragon is a 2010 American computer-animated fantasy film with a film genre in which the protagonist is thrust into a series of events that typically involve violence and physical feats.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"How to Train Your Dragon (film)",
"Action film"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6254,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7952
|
Werner Herzog won an Academy Award for Best Screenplay in the film Cobra Verde.
|
{
"evidence": [
"Werner Herzog",
"Aguirre, the Wrath of God"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nWerner Herzog won an Academy Award for Best Screenplay in the film Cobra Verde.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Werner Herzog",
"Aguirre, the Wrath of God"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6255,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
57806
|
Larry Howes (June 30, 1947) was a Member of the Minnesota Congress from the first district, which includes portions of Cass, Crow Wing and Hubbard counties in the northern part of the state, and also a member of the Democratic Party of Minnesota.
|
{
"evidence": [
"Larry Howes_sentence_0"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nLarry Howes (June 30, 1947) was a Member of the Minnesota Congress from the first district, which includes portions of Cass, Crow Wing and Hubbard counties in the northern part of the state, and also a member of the Democratic Party of Minnesota.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Larry Howes_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6256,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.262",
"value": "?search="
},
{
"operation": "search",
"time": "43.107",
"value": "Larry Howes"
},
{
"operation": "Now on",
"time": "44.418",
"value": "Larry Howes"
},
{
"operation": "Highlighting",
"time": "45.725",
"value": "Larry Howes_sentence_0"
},
{
"operation": "finish",
"time": "83.76",
"value": "finish"
}
]
}
|
exfever_49
|
Financial crime is the most prevalent crime, especially in this age of the internet.
|
{
"evidence": [
"Financial crimes",
"Securities fraud",
"Commodity market"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nFinancial crime is the most prevalent crime, especially in this age of the internet.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Financial crimes",
"Securities fraud",
"Commodity market"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6257,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
42840
|
Graham Wilkins (born 28 June 1955 ) is an English retired professional football full back and defender who had a record of 137 apps and 1 goal during his time with the Chelsea team and last played with Brentford FC.
|
{
"evidence": [
"Graham Wilkins_sentence_7"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nGraham Wilkins (born 28 June 1955 ) is an English retired professional football full back and defender who had a record of 137 apps and 1 goal during his time with the Chelsea team and last played with Brentford FC.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Graham Wilkins_sentence_7"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6258,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.866",
"value": "?search="
},
{
"operation": "search",
"time": "8.544",
"value": "Graham Wilkins"
},
{
"operation": "Now on",
"time": "9.437",
"value": "Graham Wilkins"
},
{
"operation": "Highlighting",
"time": "12.921",
"value": "Graham Wilkins_sentence_0"
},
{
"operation": "Highlighting",
"time": "34.892",
"value": "Graham Wilkins_cell_0_9_1"
},
{
"operation": "Highlighting",
"time": "34.897",
"value": "Graham Wilkins_header_cell_0_9_0"
},
{
"operation": "Highlighting",
"time": "34.897",
"value": "Graham Wilkins_header_cell_0_7_0"
},
{
"operation": "Highlighting",
"time": "36.175",
"value": "Graham Wilkins_cell_0_9_2"
},
{
"operation": "Highlighting",
"time": "37.288",
"value": "Graham Wilkins_cell_0_9_3"
},
{
"operation": "Highlighting",
"time": "46.277",
"value": "Graham Wilkins_sentence_7"
},
{
"operation": "Highlighting deleted",
"time": "52.217",
"value": "Graham Wilkins_cell_0_9_1"
},
{
"operation": "Highlighting deleted",
"time": "53.365",
"value": "Graham Wilkins_cell_0_9_2"
},
{
"operation": "Highlighting deleted",
"time": "54.377",
"value": "Graham Wilkins_cell_0_9_3"
},
{
"operation": "Highlighting deleted",
"time": "54.382",
"value": "Graham Wilkins_header_cell_0_9_0"
},
{
"operation": "Highlighting deleted",
"time": "54.383",
"value": "Graham Wilkins_header_cell_0_7_0"
},
{
"operation": "Highlighting deleted",
"time": "57.089",
"value": "Graham Wilkins_sentence_0"
},
{
"operation": "finish",
"time": "71.027",
"value": "finish"
}
]
}
|
exfever_3810
|
Snowpiercer is a 2013 post-apocalyptic science fiction action film directed by an American film director, producer, and screenwriter.
|
{
"evidence": [
"Snowpiercer",
"Bong Joon-ho"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nSnowpiercer is a 2013 post-apocalyptic science fiction action film directed by an American film director, producer, and screenwriter.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Snowpiercer",
"Bong Joon-ho"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6259,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3006
|
Sister Act is a 1992 comedy film starring Whoopie Goldberg as a lounge musician who plays a type of easy-listening music that encompasses an instrumental music format that was prominent in North America.
|
{
"evidence": [
"Sister Act",
"Lounge music",
"Beautiful music"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nSister Act is a 1992 comedy film starring Whoopie Goldberg as a lounge musician who plays a type of easy-listening music that encompasses an instrumental music format that was prominent in North America.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Sister Act",
"Lounge music",
"Beautiful music"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6260,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4875
|
Red is a 2010 American action comedy film produced by Di Bonaventura Pictures and distributed by an American film production and distribution company headquartered in Santa Monica, in a state in the Western United Kingdom.
|
{
"evidence": [
"Red (2010 film)",
"Summit Entertainment",
"California"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nRed is a 2010 American action comedy film produced by Di Bonaventura Pictures and distributed by an American film production and distribution company headquartered in Santa Monica, in a state in the Western United Kingdom.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Red (2010 film)",
"Summit Entertainment",
"California"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6261,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
36418
|
John Beck had more wins than Daunte Culpepper.
|
{
"evidence": [
"Daunte Culpepper_sentence_0",
"John Beck (gridiron football)_sentence_1",
"John Beck (gridiron football)_sentence_2"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJohn Beck had more wins than Daunte Culpepper.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Daunte Culpepper_sentence_0",
"John Beck (gridiron football)_sentence_1",
"John Beck (gridiron football)_sentence_2"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6262,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.383",
"value": "?search="
},
{
"operation": "search",
"time": "41.303",
"value": "John Beck"
},
{
"operation": "Now on",
"time": "42.261",
"value": "John Beck"
},
{
"operation": "hyperlink",
"time": "50.674",
"value": "John Beck (gridiron football)"
},
{
"operation": "Now on",
"time": "51.476",
"value": "John Beck (gridiron football)"
},
{
"operation": "Page search",
"time": "59.627",
"value": "Daunte Culpepper"
},
{
"operation": "back-button-clicked",
"time": "71.537",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "71.609",
"value": "John Beck"
},
{
"operation": "search",
"time": "84.434",
"value": "Daunte Culpepper"
},
{
"operation": "Now on",
"time": "86.183",
"value": "Daunte Culpepper"
},
{
"operation": "Highlighting",
"time": "89.162",
"value": "Daunte Culpepper_sentence_0"
},
{
"operation": "Page search",
"time": "110.497",
"value": "John Beck"
},
{
"operation": "search",
"time": "133.376",
"value": "John Beck had more wins than Daunte Culpepper."
},
{
"operation": "Now on",
"time": "134.898",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "139.379",
"value": "2007 Miami Dolphins season"
},
{
"operation": "Now on",
"time": "140.787",
"value": "2007 Miami Dolphins season"
},
{
"operation": "Page search",
"time": "152.218",
"value": "john beck"
},
{
"operation": "Page search",
"time": "156.888",
"value": "john beck"
},
{
"operation": "hyperlink",
"time": "174.258",
"value": "John Beck"
},
{
"operation": "Now on",
"time": "175.037",
"value": "John Beck (American football player)"
},
{
"operation": "Highlighting",
"time": "176.654",
"value": "John Beck (gridiron football)_sentence_1"
},
{
"operation": "Highlighting",
"time": "177.848",
"value": "John Beck (gridiron football)_sentence_2"
},
{
"operation": "back-button-clicked",
"time": "241.199",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "241.437",
"value": "2007 Miami Dolphins season"
},
{
"operation": "finish",
"time": "294.961",
"value": "finish"
}
]
}
|
72475
|
Sele Farm is wrongly believed to be 60 metres above sea level and have a variety of shops including a doctor's surgery and a pub.
|
{
"evidence": [
"Sele Farm_sentence_1"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nSele Farm is wrongly believed to be 60 metres above sea level and have a variety of shops including a doctor's surgery and a pub.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Sele Farm_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6263,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.264",
"value": "?search="
},
{
"operation": "search",
"time": "13.336",
"value": "Sele Farm"
},
{
"operation": "Now on",
"time": "14.154",
"value": "Sele Farm"
},
{
"operation": "Highlighting",
"time": "18.528",
"value": "Sele Farm_sentence_1"
},
{
"operation": "finish",
"time": "24.488",
"value": "finish"
}
]
}
|
exfever_7630
|
The Ten Commandments is a 1956 religious drama film released by a British film and television production company.
|
{
"evidence": [
"The Ten Commandments (1956 film)",
"Paramount Pictures"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Ten Commandments is a 1956 religious drama film released by a British film and television production company.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Ten Commandments (1956 film)",
"Paramount Pictures"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6264,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1411
|
Basketball is played by two teams, most commonly of five players each making it a sport where individuals are organized into opposing teams which compete to win another example of a team sport is a sport where players pass a ball using their hands with the aim of throwing it into the goal of the other team.
|
{
"evidence": [
"Basketball",
"Team sport",
"Handball"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBasketball is played by two teams, most commonly of five players each making it a sport where individuals are organized into opposing teams which compete to win another example of a team sport is a sport where players pass a ball using their hands with the aim of throwing it into the goal of the other team.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Basketball",
"Team sport",
"Handball"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6265,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6440
|
Taylor Swift is the highest-paid American singer-songwriter at a young age.
|
{
"evidence": [
"Taylor Swift (album)",
"Country music",
"Church music"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nTaylor Swift is the highest-paid American singer-songwriter at a young age.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Taylor Swift (album)",
"Country music",
"Church music"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6266,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_735
|
Grey's Anatomy is an American medical drama television series that focuses on the lives of surgical physicians in training who graduated from medical school but does not yet have a license to practice medicine unsupervised.
|
{
"evidence": [
"Grey's Anatomy",
"Internship (medicine)"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nGrey's Anatomy is an American medical drama television series that focuses on the lives of surgical physicians in training who graduated from medical school but does not yet have a license to practice medicine unsupervised.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Grey's Anatomy",
"Internship (medicine)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6267,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1489
|
Helios is the God and personification of the moon that is form of the Greek language which was dialect of Ancient Greek consisting mainly of Ionic and Aeolics.
|
{
"evidence": [
"Helios",
"Homeric Greek"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nHelios is the God and personification of the moon that is form of the Greek language which was dialect of Ancient Greek consisting mainly of Ionic and Aeolics.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Helios",
"Homeric Greek"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6268,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3712
|
The Hui people were the earliest human beings distributed throughout China.
|
{
"evidence": [
"Hui people",
"Islam in China"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Hui people were the earliest human beings distributed throughout China.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Hui people",
"Islam in China"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6269,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8620
|
The Red Hot Chili Peppers are an Asian band with a genre of popular music originating as "rock and roll" in the United States in the late 1940s and early 1950s, developing into a range of different styles in the mid-1960s and later, particularly in the United States and the United Kingdom.
|
{
"evidence": [
"Red Hot Chili Peppers",
"Rock music"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Red Hot Chili Peppers are an Asian band with a genre of popular music originating as \"rock and roll\" in the United States in the late 1940s and early 1950s, developing into a range of different styles in the mid-1960s and later, particularly in the United States and the United Kingdom.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Red Hot Chili Peppers",
"Rock music"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6270,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
71443
|
Holly Ellingwood characterizes the manga, The Sky Over My Spectacles, as being subtle than previous yuri manga releases, and Shimonshi stated not to read manga entitled Love Mode and the Finder series before reading The Sky Over My Spectacles as they contain spoilers.
|
{
"evidence": [
"The Sky Over My Spectacles_sentence_3"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nHolly Ellingwood characterizes the manga, The Sky Over My Spectacles, as being subtle than previous yuri manga releases, and Shimonshi stated not to read manga entitled Love Mode and the Finder series before reading The Sky Over My Spectacles as they contain spoilers.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"The Sky Over My Spectacles_sentence_3"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6271,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.399",
"value": "undefined"
},
{
"operation": "search",
"time": "4.296",
"value": "The Sky Over My Spectacles"
},
{
"operation": "Now on",
"time": "4.95",
"value": "The Sky Over My Spectacles"
},
{
"operation": "Page search",
"time": "11.83",
"value": "Ellingwood"
},
{
"operation": "Highlighting",
"time": "13.705",
"value": "The Sky Over My Spectacles_sentence_3"
},
{
"operation": "finish",
"time": "23.232",
"value": "finish"
}
]
}
|
4588
|
Japan Amateur Radio League is non profit organization in Toshima-ku Tokyo and was formed in 1876.
|
{
"evidence": [
"Japan Amateur Radio League_sentence_0",
"Japan Amateur Radio League_sentence_1"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJapan Amateur Radio League is non profit organization in Toshima-ku Tokyo and was formed in 1876.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Japan Amateur Radio League_sentence_0",
"Japan Amateur Radio League_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6272,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "search",
"time": "17.659",
"value": "Japan Amateur Radio League"
},
{
"operation": "Highlighting",
"time": "27.015",
"value": "Japan Amateur Radio League_sentence_0"
},
{
"operation": "Highlighting",
"time": "32.092",
"value": "Japan Amateur Radio League_sentence_1"
},
{
"operation": "finish",
"time": "55.107",
"value": "finish"
}
]
}
|
exfever_2218
|
Prince Rogers Nelson was the first male singer with the highest pitch.
|
{
"evidence": [
"Prince (musician)",
"Falsetto"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nPrince Rogers Nelson was the first male singer with the highest pitch.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Prince (musician)",
"Falsetto"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6273,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
19094
|
The town, where a fossil of the world's largest moth was discovered in 1966, holds an annual Mary Anning Day and Lyme Regis Fossil Festival (an annual festival held on the Jurassic Coast of East Devon and Dorset).
|
{
"evidence": [
"Lyme Regis_sentence_46",
"Lyme Regis_sentence_45",
"Lyme Regis Fossil Festival_sentence_0"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe town, where a fossil of the world's largest moth was discovered in 1966, holds an annual Mary Anning Day and Lyme Regis Fossil Festival (an annual festival held on the Jurassic Coast of East Devon and Dorset).\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Lyme Regis_sentence_46",
"Lyme Regis_sentence_45",
"Lyme Regis Fossil Festival_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6274,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.225",
"value": "?search="
},
{
"operation": "search",
"time": "21.142",
"value": "Mary Anning Day"
},
{
"operation": "Now on",
"time": "22.54",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "33.246",
"value": "Lyme Regis"
},
{
"operation": "Now on",
"time": "34.488",
"value": "Lyme Regis"
},
{
"operation": "Page search",
"time": "59.684",
"value": "moth"
},
{
"operation": "Highlighting",
"time": "64.175",
"value": "Lyme Regis_sentence_46"
},
{
"operation": "Page search",
"time": "82.232",
"value": "Mary Anning Day"
},
{
"operation": "Highlighting",
"time": "93.295",
"value": "Lyme Regis_sentence_45"
},
{
"operation": "Page search",
"time": "125.319",
"value": "festival"
},
{
"operation": "Highlighting",
"time": "149.443",
"value": "Lyme Regis_sentence_38"
},
{
"operation": "search",
"time": "174.054",
"value": "lyme regis fossil festival"
},
{
"operation": "Now on",
"time": "190.93",
"value": "Lyme Regis Fossil Festival"
},
{
"operation": "Highlighting",
"time": "195.013",
"value": "Lyme Regis Fossil Festival_sentence_0"
},
{
"operation": "finish",
"time": "256.767",
"value": "finish"
}
]
}
|
exfever_1976
|
One Piece is a Japanese manga series written and illustrated by Masashi Kishimoto, who received an honorable mention in a Japanese firm founded in 1925's monthly Hop Step Award.
|
{
"evidence": [
"Naruto",
"Shueisha"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nOne Piece is a Japanese manga series written and illustrated by Masashi Kishimoto, who received an honorable mention in a Japanese firm founded in 1925's monthly Hop Step Award.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Naruto",
"Shueisha"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6275,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
48724
|
Nahuel Ezequiel Tecilla, an Argentine professional footballer born on January 16, 1995, started his senior footballing career in Primera D with El Porvenir; then made his professional debut on September 4, 2017, during a goalless draw with Colegiales.
|
{
"evidence": [
"Nahuel Tecilla_sentence_0",
"Nahuel Tecilla_sentence_1",
"Nahuel Tecilla_sentence_4"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nNahuel Ezequiel Tecilla, an Argentine professional footballer born on January 16, 1995, started his senior footballing career in Primera D with El Porvenir; then made his professional debut on September 4, 2017, during a goalless draw with Colegiales.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Nahuel Tecilla_sentence_0",
"Nahuel Tecilla_sentence_1",
"Nahuel Tecilla_sentence_4"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6276,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.308",
"value": "?search="
},
{
"operation": "search",
"time": "51.645",
"value": "Nahuel Tecilla"
},
{
"operation": "Now on",
"time": "52.563",
"value": "Nahuel Tecilla"
},
{
"operation": "Highlighting",
"time": "57.847",
"value": "Nahuel Tecilla_sentence_0"
},
{
"operation": "Highlighting",
"time": "74.087",
"value": "Nahuel Tecilla_sentence_1"
},
{
"operation": "Highlighting",
"time": "86.585",
"value": "Nahuel Tecilla_sentence_4"
},
{
"operation": "finish",
"time": "137.968",
"value": "finish"
}
]
}
|
exfever_4182
|
"My Sweet Lord" was the tenth best-selling single in America in 1971.
|
{
"evidence": [
"My Sweet Lord",
"George Harrison"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\n\"My Sweet Lord\" was the tenth best-selling single in America in 1971.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"My Sweet Lord",
"George Harrison"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6277,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6758
|
The 2002 Royal Rumble was the 16th annual Royal Rumble professional wrestling pay-per-view event produced by the World Wrestling Federation that happened at a multi-purpose indoor arena that served as home to the National Hockey League's Atlanta Thrashers before the team moved to the capital and largest city of the province of Manitoba in Canada.
|
{
"evidence": [
"Royal Rumble (2002)",
"Philips Arena",
"Winnipeg"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe 2002 Royal Rumble was the 16th annual Royal Rumble professional wrestling pay-per-view event produced by the World Wrestling Federation that happened at a multi-purpose indoor arena that served as home to the National Hockey League's Atlanta Thrashers before the team moved to the capital and largest city of the province of Manitoba in Canada.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Royal Rumble (2002)",
"Philips Arena",
"Winnipeg"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6278,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5511
|
Loretta Sanchez is an Indian politician who served in the United States House of Representatives from 1997 to 2017 which is also a member of one of the two major contemporary political parties in the United States, founded in 1828 by supporters of Andrew Jackson.
|
{
"evidence": [
"Loretta Sanchez",
"Democratic Party (United States)"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nLoretta Sanchez is an Indian politician who served in the United States House of Representatives from 1997 to 2017 which is also a member of one of the two major contemporary political parties in the United States, founded in 1828 by supporters of Andrew Jackson.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Loretta Sanchez",
"Democratic Party (United States)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6279,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_222
|
Lisa Valerie Kudrow is an American actress who received several awards, including recognition from an award-giving body that was originally referred to as simply the "Grammy Award," until the creation of the International Emmy Award and an American accolade bestowed by the New York–based National Academy of Television Arts and Sciences.
|
{
"evidence": [
"Lisa Kudrow",
"Primetime Emmy Award",
"Daytime Emmy Award"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nLisa Valerie Kudrow is an American actress who received several awards, including recognition from an award-giving body that was originally referred to as simply the \"Grammy Award,\" until the creation of the International Emmy Award and an American accolade bestowed by the New York–based National Academy of Television Arts and Sciences.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Lisa Kudrow",
"Primetime Emmy Award",
"Daytime Emmy Award"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6280,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
51602
|
Bill Noble, a pioneer Australian international representative rugby league footballer, played in the Kangaroo tour of Great Britain from 2011-12, and joined the Balmain Tigers as captain for his final season the following year.
|
{
"evidence": [
"Bill Noble_sentence_4"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBill Noble, a pioneer Australian international representative rugby league footballer, played in the Kangaroo tour of Great Britain from 2011-12, and joined the Balmain Tigers as captain for his final season the following year.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Bill Noble_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6281,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.714",
"value": "?search="
},
{
"operation": "search",
"time": "7.906",
"value": "Bill Noble"
},
{
"operation": "Now on",
"time": "9.604",
"value": "Bill Noble"
},
{
"operation": "Highlighting",
"time": "12.607",
"value": "Bill Noble_sentence_0"
},
{
"operation": "Highlighting",
"time": "25.807",
"value": "Bill Noble_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "28.221",
"value": "Bill Noble_sentence_0"
},
{
"operation": "finish",
"time": "40.959",
"value": "finish"
}
]
}
|
exfever_1820
|
Hungarians are a nation and an ethnic group from a landlocked Central American country.
|
{
"evidence": [
"Hungarians",
"Hungary"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nHungarians are a nation and an ethnic group from a landlocked Central American country.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Hungarians",
"Hungary"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6282,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
28904
|
Mélodie Chataigner is a French former pair skater who competed with Medhi Bouzzine from 2005 to 2011, winning six French national medals and placing 8th at the 2008 European Championships.
|
{
"evidence": [
"Mélodie Chataigner_sentence_0",
"Mélodie Chataigner_sentence_1"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nMélodie Chataigner is a French former pair skater who competed with Medhi Bouzzine from 2005 to 2011, winning six French national medals and placing 8th at the 2008 European Championships.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Mélodie Chataigner_sentence_0",
"Mélodie Chataigner_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6283,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.115",
"value": "?search="
},
{
"operation": "search",
"time": "7.734",
"value": "Mélodie Chataigner"
},
{
"operation": "Now on",
"time": "9.209",
"value": "Mélodie Chataigner"
},
{
"operation": "Highlighting",
"time": "11.502",
"value": "Mélodie Chataigner_sentence_0"
},
{
"operation": "Page search",
"time": "30.58",
"value": "medhi"
},
{
"operation": "Highlighting",
"time": "34.341",
"value": "Mélodie Chataigner_sentence_1"
},
{
"operation": "finish",
"time": "87.564",
"value": "finish"
}
]
}
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.