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 |
|---|---|---|---|---|---|---|---|---|
exfever_4251
|
Lost Highway is a 1997 neo-noir film shot in California's largest city, extending into an urbanized valley that houses Universal Studios Hollywood theme park.
|
{
"evidence": [
"Lost Highway (film)",
"Los Angeles",
"San Fernando Valley"
],
"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:\nLost Highway is a 1997 neo-noir film shot in California's largest city, extending into an urbanized valley that houses Universal Studios Hollywood theme park.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Lost Highway (film)",
"Los Angeles",
"San Fernando Valley"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6784,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_119
|
Jordan Spence is a professional footballer who played for a professional football club based in Leyton, East Asia.
|
{
"evidence": [
"Jordan Spence",
"Leyton Orient F.C."
],
"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:\nJordan Spence is a professional footballer who played for a professional football club based in Leyton, East Asia.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Jordan Spence",
"Leyton Orient F.C."
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6785,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5924
|
Kid Rock is an American dancer whose style alternates between different styles of music, including a style that draws directly from the blues, rhythm, and blues genres of a diverse range of music and musical genres largely developed by African Americans.
|
{
"evidence": [
"Kid Rock",
"Rock music",
"African-American 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:\nKid Rock is an American dancer whose style alternates between different styles of music, including a style that draws directly from the blues, rhythm, and blues genres of a diverse range of music and musical genres largely developed by African Americans.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Kid Rock",
"Rock music",
"African-American music"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6786,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1563
|
"Game of Thrones" was created for an American pay television network.
|
{
"evidence": [
"Game of Thrones",
"HBO"
],
"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\"Game of Thrones\" was created for an American pay television network.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Game of Thrones",
"HBO"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6787,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4346
|
30 Seconds to Mars is the most sold album by American rock band Thirty Seconds to Mars.
|
{
"evidence": [
"30 Seconds to Mars (album)",
"Jared Leto"
],
"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:\n30 Seconds to Mars is the most sold album by American rock band Thirty Seconds to Mars.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"30 Seconds to Mars (album)",
"Jared Leto"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6788,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
44616
|
Prosthechea brassavolae is a plant genus classified within the family Orchidaceae, in the order Asparagales.
|
{
"evidence": [
"Prosthechea brassavolae_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:\nProsthechea brassavolae is a plant genus classified within the family Orchidaceae, in the order Asparagales.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Prosthechea brassavolae_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6789,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0",
"value": "?search="
},
{
"operation": "search",
"time": "0.009",
"value": "Prosthechea brassavolae"
},
{
"operation": "Now on",
"time": "0.011",
"value": "Prosthechea brassavolae"
},
{
"operation": "Highlighting",
"time": "0.021",
"value": "Prosthechea brassavolae_sentence_0"
},
{
"operation": "finish",
"time": "0.078",
"value": "finish"
},
{
"operation": "Now on",
"time": "1620830511.394",
"value": "Prosthechea brassavolae"
},
{
"operation": "search",
"time": "1620830520.381",
"value": "Contains...plant genus"
},
{
"operation": "Now on",
"time": "1620830521.644",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "1620830532.587",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "1620830532.912",
"value": "Prosthechea brassavolae"
},
{
"operation": "hyperlink",
"time": "1620830534.7",
"value": "Prosthechea"
},
{
"operation": "Now on",
"time": "1620830535.391",
"value": "Prosthechea"
},
{
"operation": "back-button-clicked",
"time": "1620830554.092",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "1620830554.412",
"value": "Prosthechea brassavolae"
},
{
"operation": "Highlighting deleted",
"time": "1620830563.619",
"value": "Prosthechea brassavolae_sentence_0"
},
{
"operation": "Highlighting",
"time": "1620830564.62",
"value": "Prosthechea brassavolae_cell_0_9_1"
},
{
"operation": "Highlighting",
"time": "1620830564.626",
"value": "Prosthechea brassavolae_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "1620830564.627",
"value": "Prosthechea brassavolae_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "1620830566.364",
"value": "Prosthechea brassavolae_cell_0_9_1"
},
{
"operation": "Highlighting deleted",
"time": "1620830566.372",
"value": "Prosthechea brassavolae_header_cell_0_1_0"
},
{
"operation": "Highlighting deleted",
"time": "1620830566.374",
"value": "Prosthechea brassavolae_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "1620830573.035",
"value": "Prosthechea brassavolae_cell_0_9_0"
},
{
"operation": "Highlighting",
"time": "1620830573.042",
"value": "Prosthechea brassavolae_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "1620830573.042",
"value": "Prosthechea brassavolae_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "1620830574.107",
"value": "Prosthechea brassavolae_cell_0_9_1"
},
{
"operation": "hyperlink",
"time": "1620830575.733",
"value": "Prosthechea"
},
{
"operation": "Now on",
"time": "1620830576.349",
"value": "Prosthechea"
},
{
"operation": "Page search",
"time": "1620830586.142",
"value": "brassavolae"
},
{
"operation": "back-button-clicked",
"time": "1620830609.693",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "1620830609.921",
"value": "Prosthechea brassavolae"
},
{
"operation": "Highlighting",
"time": "1620830611.79",
"value": "Prosthechea brassavolae_sentence_0"
},
{
"operation": "finish",
"time": "1620830618.071",
"value": "finish"
}
]
}
|
exfever_3887
|
Timothy David Olyphant is an American actor who appeared in the 2001 American action heist film starring Nicolas Cage.
|
{
"evidence": [
"Timothy Olyphant",
"Gone in 60 Seconds (2000 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:\nTimothy David Olyphant is an American actor who appeared in the 2001 American action heist film starring Nicolas Cage.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Timothy Olyphant",
"Gone in 60 Seconds (2000 film)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6790,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
91128
|
Immenhausen is a town in the district of Kassel, in Hesse, Germany but no known person was born here.
|
{
"evidence": [
"Immenhausen_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:\nImmenhausen is a town in the district of Kassel, in Hesse, Germany but no known person was born here.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Immenhausen_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6791,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.627",
"value": "?search="
},
{
"operation": "search",
"time": "12.464",
"value": "Immenhausen"
},
{
"operation": "Now on",
"time": "13.666",
"value": "Immenhausen"
},
{
"operation": "Highlighting",
"time": "16.181",
"value": "Immenhausen_sentence_2"
},
{
"operation": "finish",
"time": "28.08",
"value": "finish"
}
]
}
|
exfever_1995
|
Vacation is a 2015 American road comedy film starring an Australian actor who appeared in an American science fiction action film directed by J. J. Abrams.
|
{
"evidence": [
"Vacation (2015 film)",
"Chris Hemsworth",
"Star Trek (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:\nVacation is a 2015 American road comedy film starring an Australian actor who appeared in an American science fiction action film directed by J. J. Abrams.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Vacation (2015 film)",
"Chris Hemsworth",
"Star Trek (film)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6792,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3339
|
The Conjuring is a 2013 horror film written by an American screenwriter who co-wrote a 2007 supernatural horror-thriller film, starring Hilary Swank.
|
{
"evidence": [
"The Conjuring",
"Chad Hayes (writer)",
"The Reaping"
],
"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 Conjuring is a 2013 horror film written by an American screenwriter who co-wrote a 2007 supernatural horror-thriller film, starring Hilary Swank.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Conjuring",
"Chad Hayes (writer)",
"The Reaping"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6793,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
19431
|
The 1382 Dover Straits earthquake which had an estimated magnitude of 6.0 Ms produced significant aftershocks on 23 May and on 24 May, the latter being the strongest recorded and was also described as a "waterquake" affecting anchored ships.
|
{
"evidence": [
"1382 Dover Straits earthquake_sentence_1",
"1382 Dover Straits earthquake_sentence_12",
"1382 Dover Straits earthquake_sentence_13"
],
"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 1382 Dover Straits earthquake which had an estimated magnitude of 6.0 Ms produced significant aftershocks on 23 May and on 24 May, the latter being the strongest recorded and was also described as a \"waterquake\" affecting anchored ships.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"1382 Dover Straits earthquake_sentence_1",
"1382 Dover Straits earthquake_sentence_12",
"1382 Dover Straits earthquake_sentence_13"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6794,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.486",
"value": "?search="
},
{
"operation": "search",
"time": "17.123",
"value": "1382 Dover Straits earthquake"
},
{
"operation": "Now on",
"time": "17.835",
"value": "1382 Dover Straits earthquake"
},
{
"operation": "Highlighting",
"time": "23.014",
"value": "1382 Dover Straits earthquake_sentence_1"
},
{
"operation": "Highlighting",
"time": "35.125",
"value": "1382 Dover Straits earthquake_sentence_12"
},
{
"operation": "Highlighting",
"time": "43.296",
"value": "1382 Dover Straits earthquake_sentence_13"
},
{
"operation": "finish",
"time": "66.564",
"value": "finish"
}
]
}
|
70916
|
Monte Westmore was Italian actor Rudolph Valentino's makeup artist until the actor's death in 1926 and helped Kurt Warner stop drinking.
|
{
"evidence": [
"Monte Westmore_sentence_10",
"Monte Westmore_sentence_11",
"Monte Westmore_sentence_14"
],
"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:\nMonte Westmore was Italian actor Rudolph Valentino's makeup artist until the actor's death in 1926 and helped Kurt Warner stop drinking.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Monte Westmore_sentence_10",
"Monte Westmore_sentence_11",
"Monte Westmore_sentence_14"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6795,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.847",
"value": "?search="
},
{
"operation": "search",
"time": "8.751",
"value": "Monte Westmore"
},
{
"operation": "Now on",
"time": "9.776",
"value": "Monte Westmore"
},
{
"operation": "Highlighting",
"time": "185.82",
"value": "Monte Westmore_sentence_2"
},
{
"operation": "Highlighting",
"time": "515.71",
"value": "Monte Westmore_sentence_3"
},
{
"operation": "hyperlink",
"time": "524.846",
"value": "Gone with the Wind"
},
{
"operation": "Now on",
"time": "526.549",
"value": "Gone with the Wind (film)"
},
{
"operation": "back-button-clicked",
"time": "527.431",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "527.491",
"value": "Monte Westmore"
},
{
"operation": "Highlighting deleted",
"time": "528.786",
"value": "Monte Westmore_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "529.313",
"value": "Monte Westmore_sentence_3"
},
{
"operation": "Highlighting",
"time": "545.715",
"value": "Monte Westmore_sentence_10"
},
{
"operation": "Highlighting",
"time": "547.044",
"value": "Monte Westmore_sentence_11"
},
{
"operation": "Page search",
"time": "571.823",
"value": "kurt warner"
},
{
"operation": "hyperlink",
"time": "589.253",
"value": "H. Monte Westmore_sentence_12\n\nB. Warner"
},
{
"operation": "Now on",
"time": "589.88",
"value": "H. B. Warner"
},
{
"operation": "Page search",
"time": "597.103",
"value": "Monte Westmore"
},
{
"operation": "back-button-clicked",
"time": "606.898",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "606.967",
"value": "Monte Westmore"
},
{
"operation": "Highlighting",
"time": "609.267",
"value": "Monte Westmore_sentence_14"
},
{
"operation": "Highlighting",
"time": "610.09",
"value": "Monte Westmore_sentence_12Monte Westmore_sentence_13"
},
{
"operation": "Highlighting deleted",
"time": "613.217",
"value": "Monte Westmore_sentence_12Monte Westmore_sentence_13"
},
{
"operation": "hyperlink",
"time": "625.658",
"value": "H. Monte Westmore_sentence_12\n\nB. Warner"
},
{
"operation": "Now on",
"time": "626.059",
"value": "H. B. Warner"
},
{
"operation": "back-button-clicked",
"time": "635.676",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "635.74",
"value": "Monte Westmore"
},
{
"operation": "Highlighting deleted",
"time": "636.538",
"value": "Monte Westmore_sentence_11"
},
{
"operation": "Highlighting deleted",
"time": "637.2",
"value": "Monte Westmore_sentence_10"
},
{
"operation": "Highlighting deleted",
"time": "652.747",
"value": "Monte Westmore_sentence_14"
},
{
"operation": "Highlighting",
"time": "654.108",
"value": "Monte Westmore_sentence_10"
},
{
"operation": "Highlighting",
"time": "655.156",
"value": "Monte Westmore_sentence_11"
},
{
"operation": "Highlighting",
"time": "656.539",
"value": "Monte Westmore_sentence_14"
},
{
"operation": "finish",
"time": "660.869",
"value": "finish"
}
]
}
|
69619
|
Chinary Ung is a composer who won the University of Louisville Grawemeyer Award for musical composition and the Kennedy Center Friedheim Award.
|
{
"evidence": [
"Chinary Ung_sentence_0",
"Chinary Ung_sentence_2",
"Chinary Ung_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:\nChinary Ung is a composer who won the University of Louisville Grawemeyer Award for musical composition and the Kennedy Center Friedheim Award.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Chinary Ung_sentence_0",
"Chinary Ung_sentence_2",
"Chinary Ung_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6796,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "4.173",
"value": "undefined"
},
{
"operation": "search",
"time": "17.188",
"value": "Chinary Ung"
},
{
"operation": "Now on",
"time": "23.775",
"value": "Chinary Ung"
},
{
"operation": "Page search",
"time": "40.814",
"value": "Grawemeyer"
},
{
"operation": "Highlighting",
"time": "44.724",
"value": "Chinary Ung_sentence_2"
},
{
"operation": "Page search",
"time": "57.655",
"value": "Kennedy"
},
{
"operation": "Highlighting",
"time": "60.893",
"value": "Chinary Ung_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "117.004",
"value": "Chinary Ung_sentence_2"
},
{
"operation": "Highlighting",
"time": "117.158",
"value": "Chinary Ung_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "118.242",
"value": "Chinary Ung_sentence_3"
},
{
"operation": "Highlighting",
"time": "118.42",
"value": "Chinary Ung_sentence_3"
},
{
"operation": "Highlighting",
"time": "126.099",
"value": "Chinary Ung_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "127.258",
"value": "Chinary Ung_sentence_2"
},
{
"operation": "Highlighting",
"time": "127.449",
"value": "Chinary Ung_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "128.029",
"value": "Chinary Ung_sentence_3"
},
{
"operation": "Highlighting",
"time": "128.214",
"value": "Chinary Ung_sentence_3"
},
{
"operation": "finish",
"time": "143.838",
"value": "finish"
}
]
}
|
21755
|
Gymnopilus naucorioides is a species of mushroom, with a cap of two to five centimetres, which grows on hemlock (aka Tsuga)which is a conifer that is perceived to smell like the unrelated poison hemlock.
|
{
"evidence": [
"Gymnopilus naucorioides_sentence_0",
"Gymnopilus naucorioides_sentence_1",
"Gymnopilus naucorioides_sentence_2",
"Tsuga_sentence_2",
"Tsuga_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:\nGymnopilus naucorioides is a species of mushroom, with a cap of two to five centimetres, which grows on hemlock (aka Tsuga)which is a conifer that is perceived to smell like the unrelated poison hemlock.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Gymnopilus naucorioides_sentence_0",
"Gymnopilus naucorioides_sentence_1",
"Gymnopilus naucorioides_sentence_2",
"Tsuga_sentence_2",
"Tsuga_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6797,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.601",
"value": "?search="
},
{
"operation": "search",
"time": "10.385",
"value": "Gymnopilus naucorioides"
},
{
"operation": "Now on",
"time": "11.068",
"value": "Gymnopilus naucorioides"
},
{
"operation": "Highlighting",
"time": "22.045",
"value": "Gymnopilus naucorioides_sentence_0"
},
{
"operation": "Highlighting",
"time": "24.677",
"value": "Gymnopilus naucorioides_sentence_1"
},
{
"operation": "Highlighting",
"time": "27.356",
"value": "Gymnopilus naucorioides_sentence_2"
},
{
"operation": "hyperlink",
"time": "45.094",
"value": "hemlock"
},
{
"operation": "Now on",
"time": "45.725",
"value": "Tsuga"
},
{
"operation": "Highlighting",
"time": "47.638",
"value": "Tsuga_sentence_2"
},
{
"operation": "back-button-clicked",
"time": "50.847",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "50.948",
"value": "Gymnopilus naucorioides"
},
{
"operation": "hyperlink",
"time": "56.348",
"value": "hemlock"
},
{
"operation": "Now on",
"time": "56.79",
"value": "Tsuga"
},
{
"operation": "Highlighting",
"time": "72.995",
"value": "Tsuga_sentence_3"
},
{
"operation": "finish",
"time": "79.063",
"value": "finish"
}
]
}
|
exfever_5826
|
Kingdom Hearts III is the second action role-playing game developed and published for the PlayStation 4.
|
{
"evidence": [
"Kingdom Hearts III",
"Square Enix"
],
"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:\nKingdom Hearts III is the second action role-playing game developed and published for the PlayStation 4.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Kingdom Hearts III",
"Square Enix"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6798,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
18102
|
During the Russian Civil War, Konstantin Chebotaryov got enlisted in the White Army; later on, he fled to the east after being crushed by the Reds.
|
{
"evidence": [
"Konstantin Chebotaryov_sentence_12",
"Konstantin Chebotaryov_sentence_13"
],
"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:\nDuring the Russian Civil War, Konstantin Chebotaryov got enlisted in the White Army; later on, he fled to the east after being crushed by the Reds.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Konstantin Chebotaryov_sentence_12",
"Konstantin Chebotaryov_sentence_13"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6799,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.583",
"value": "undefined"
},
{
"operation": "search",
"time": "15.291",
"value": "Konstantin Chebotaryov"
},
{
"operation": "Now on",
"time": "16.187",
"value": "Konstantin Chebotaryov"
},
{
"operation": "Page search",
"time": "30.209",
"value": "white army"
},
{
"operation": "Highlighting",
"time": "35.089",
"value": "Konstantin Chebotaryov_sentence_12"
},
{
"operation": "Highlighting",
"time": "43.211",
"value": "Konstantin Chebotaryov_sentence_13"
},
{
"operation": "finish",
"time": "48.123",
"value": "finish"
}
]
}
|
exfever_5697
|
Gone with the Wind is a novel that shows how disastrous the "March to the Sea" commanded by a general in the Union Army during the civil war in the United Kingdom between the Union and the Confederacy.
|
{
"evidence": [
"Gone with the Wind (novel)",
"William Tecumseh Sherman",
"American Civil War"
],
"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:\nGone with the Wind is a novel that shows how disastrous the \"March to the Sea\" commanded by a general in the Union Army during the civil war in the United Kingdom between the Union and the Confederacy.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Gone with the Wind (novel)",
"William Tecumseh Sherman",
"American Civil War"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6800,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4719
|
Deadpool is Ed Skrein's tenth film.
|
{
"evidence": [
"Deadpool (film)",
"Tim Miller (director)",
"Film director"
],
"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:\nDeadpool is Ed Skrein's tenth film.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Deadpool (film)",
"Tim Miller (director)",
"Film director"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6801,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3117
|
Betty Buckley won the 1983 Oscar Award for Best Featured Actress in a Musical for her role as Grizabella in the original Broadway production of a sung-through musical composed by Andrew Lloyd Webber.
|
{
"evidence": [
"Betty Buckley",
"Cats (musical)"
],
"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:\nBetty Buckley won the 1983 Oscar Award for Best Featured Actress in a Musical for her role as Grizabella in the original Broadway production of a sung-through musical composed by Andrew Lloyd Webber.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Betty Buckley",
"Cats (musical)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6802,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
50099
|
Winfried Bühler worked on the reconstruction of the Bonn University buildings in North Rhine-Westphalia in Germany, guaranteeing him the chance to enroll for the Winter Term.
|
{
"evidence": [
"Winfried Bühler_sentence_7",
"University of Bonn_sentence_0",
"Winfried Bühler_sentence_8"
],
"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:\nWinfried Bühler worked on the reconstruction of the Bonn University buildings in North Rhine-Westphalia in Germany, guaranteeing him the chance to enroll for the Winter Term.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Winfried Bühler_sentence_7",
"University of Bonn_sentence_0",
"Winfried Bühler_sentence_8"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6803,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.393",
"value": "?search="
},
{
"operation": "search",
"time": "13.541",
"value": "Winfried Bühler"
},
{
"operation": "Now on",
"time": "14.535",
"value": "Winfried Bühler"
},
{
"operation": "Page search",
"time": "27.198",
"value": "Winter"
},
{
"operation": "Highlighting",
"time": "32.201",
"value": "Winfried Bühler_sentence_7"
},
{
"operation": "hyperlink",
"time": "41.159",
"value": "Bonn University"
},
{
"operation": "Now on",
"time": "43.517",
"value": "University of Bonn"
},
{
"operation": "Highlighting",
"time": "48.223",
"value": "University of Bonn_sentence_0"
},
{
"operation": "back-button-clicked",
"time": "50.849",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "51.179",
"value": "Winfried Bühler"
},
{
"operation": "Highlighting",
"time": "53.224",
"value": "Winfried Bühler_sentence_8"
},
{
"operation": "finish",
"time": "61.238",
"value": "finish"
}
]
}
|
10347
|
Tasmania has a climate similar to places like the United Kingdom, New Zealand, and the Pacific Northwest region of the United States, and Canada, it was heavily influenced by the Roaring Forties with strong westerly winds in the southern hemisphere, which cool the rising air mass and lower air temperatures in this region.
|
{
"evidence": [
"Climate of Tasmania_sentence_8"
],
"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:\nTasmania has a climate similar to places like the United Kingdom, New Zealand, and the Pacific Northwest region of the United States, and Canada, it was heavily influenced by the Roaring Forties with strong westerly winds in the southern hemisphere, which cool the rising air mass and lower air temperatures in this region.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Climate of Tasmania_sentence_8"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6804,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.546",
"value": "?search="
},
{
"operation": "search",
"time": "36.059",
"value": "Contains...climate in tasmania"
},
{
"operation": "Now on",
"time": "37.156",
"value": "undefined"
},
{
"operation": "search",
"time": "59.301",
"value": "tasmania"
},
{
"operation": "Now on",
"time": "59.902",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "65.576",
"value": "Premier of Tasmania"
},
{
"operation": "Now on",
"time": "66.603",
"value": "Premier of Tasmania"
},
{
"operation": "back-button-clicked",
"time": "78.958",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "79.579",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "85.605",
"value": "Local government areas of Tasmania"
},
{
"operation": "Now on",
"time": "86.514",
"value": "Local government areas of Tasmania"
},
{
"operation": "hyperlink",
"time": "107.215",
"value": "Government of Tasmania"
},
{
"operation": "Now on",
"time": "107.913",
"value": "Tasmanian Government"
},
{
"operation": "search",
"time": "119.955",
"value": "tasmanian climate"
},
{
"operation": "Now on",
"time": "120.801",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "133.967",
"value": "Climate Deloraine, Tasmania_section_5"
},
{
"operation": "Now on",
"time": "134.481",
"value": "Deloraine, Tasmania#Climate Deloraine.2C Tasmania section 5"
},
{
"operation": "search",
"time": "179.576",
"value": "tasmania climate"
},
{
"operation": "Now on",
"time": "180.094",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "183.95",
"value": "Climate of Tasmania"
},
{
"operation": "Now on",
"time": "184.48",
"value": "Climate of Tasmania"
},
{
"operation": "Highlighting",
"time": "195.004",
"value": "Climate of Tasmania_sentence_8"
},
{
"operation": "Page search",
"time": "213.977",
"value": "roaring"
},
{
"operation": "Page search",
"time": "280.657",
"value": "westerly"
},
{
"operation": "hyperlink",
"time": "311.403",
"value": "8 See also Climate of Tasmania_section_13"
},
{
"operation": "finish",
"time": "349.585",
"value": "finish"
}
]
}
|
exfever_4024
|
The Little Prince is a 2015 film directed by a Turkish film director, writer, producer, and animator.
|
{
"evidence": [
"The Little Prince (2015 film)",
"Mark Osborne (filmmaker)"
],
"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 Little Prince is a 2015 film directed by a Turkish film director, writer, producer, and animator.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Little Prince (2015 film)",
"Mark Osborne (filmmaker)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6805,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3476
|
One of the best Bollywood movies of 2020 is Thappad, which gained 4.5 good reviews from its viewers.
|
{
"evidence": [
"Akshay Kumar",
"Bollywood"
],
"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:\nOne of the best Bollywood movies of 2020 is Thappad, which gained 4.5 good reviews from its viewers.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Akshay Kumar",
"Bollywood"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6806,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
35780
|
South African Tony Maggs finished first in the 1962 French Grand Prix with three points ahead of the second placer.
|
{
"evidence": [
"Tony Maggs_sentence_5"
],
"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:\nSouth African Tony Maggs finished first in the 1962 French Grand Prix with three points ahead of the second placer.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Tony Maggs_sentence_5"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6807,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.725",
"value": "?search="
},
{
"operation": "search",
"time": "8.039",
"value": "Tony Magga"
},
{
"operation": "Now on",
"time": "8.768",
"value": "undefined"
},
{
"operation": "search",
"time": "16.52",
"value": "Tony maggs"
},
{
"operation": "Now on",
"time": "17.366",
"value": "Tony Maggs"
},
{
"operation": "Highlighting",
"time": "34.675",
"value": "Tony Maggs_sentence_5"
},
{
"operation": "finish",
"time": "52.759",
"value": "finish"
}
]
}
|
37023
|
Casey Matthews did nine tackles three times and finished as a teams sixth leading tackler, College football news (Online podcast) says he is someone to know.
|
{
"evidence": [
"Casey Matthews_sentence_19",
"Casey Matthews_sentence_17",
"Casey Matthews_sentence_18"
],
"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:\nCasey Matthews did nine tackles three times and finished as a teams sixth leading tackler, College football news (Online podcast) says he is someone to know.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Casey Matthews_sentence_19",
"Casey Matthews_sentence_17",
"Casey Matthews_sentence_18"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6808,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.76",
"value": "?search="
},
{
"operation": "Now on",
"time": "25.899",
"value": "?search="
},
{
"operation": "search",
"time": "32.389",
"value": "Casey Matthews"
},
{
"operation": "Now on",
"time": "33.093",
"value": "Casey Matthews"
},
{
"operation": "Page search",
"time": "84.46",
"value": "someone"
},
{
"operation": "Page search",
"time": "110.253",
"value": "college football news"
},
{
"operation": "Highlighting",
"time": "119.427",
"value": "Casey Matthews_sentence_19"
},
{
"operation": "Highlighting",
"time": "127.831",
"value": "Casey Matthews_sentence_17"
},
{
"operation": "Highlighting",
"time": "130.527",
"value": "Casey Matthews_sentence_18"
},
{
"operation": "finish",
"time": "135.575",
"value": "finish"
}
]
}
|
56223
|
Italian rider Nello Pagani won the 500 cc race and the 125 cc race at the 1949 Dutch TT, the third race of the 1949 Motorcycle Grand Prix season, which took place on the weekend of 9 July 1949 at the Assen circuit.
|
{
"evidence": [
"1949 Dutch TT_sentence_2",
"1949 Dutch TT_sentence_3",
"1949 Dutch TT_sentence_0",
"1949 Dutch TT_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:\nItalian rider Nello Pagani won the 500 cc race and the 125 cc race at the 1949 Dutch TT, the third race of the 1949 Motorcycle Grand Prix season, which took place on the weekend of 9 July 1949 at the Assen circuit.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"1949 Dutch TT_sentence_2",
"1949 Dutch TT_sentence_3",
"1949 Dutch TT_sentence_0",
"1949 Dutch TT_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6809,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.118",
"value": "?search="
},
{
"operation": "search",
"time": "22.359",
"value": "Contains...1949 Motorcycle Grand Prix season"
},
{
"operation": "Now on",
"time": "24.022",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "33.334",
"value": "1949 Dutch TT"
},
{
"operation": "Now on",
"time": "34.259",
"value": "1949 Dutch TT"
},
{
"operation": "Highlighting",
"time": "46.174",
"value": "1949 Dutch TT_sentence_2"
},
{
"operation": "Highlighting",
"time": "51.011",
"value": "1949 Dutch TT_sentence_3"
},
{
"operation": "Highlighting",
"time": "58.13",
"value": "1949 Dutch TT_sentence_0"
},
{
"operation": "Highlighting",
"time": "61.064",
"value": "1949 Dutch TT_sentence_1"
},
{
"operation": "finish",
"time": "69.508",
"value": "finish"
}
]
}
|
exfever_4680
|
iZombie is a supernatural procedural crime drama television series developed for an American English-language commercial broadcast television network.
|
{
"evidence": [
"IZombie (TV series)",
"The CW"
],
"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:\niZombie is a supernatural procedural crime drama television series developed for an American English-language commercial broadcast television network.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"IZombie (TV series)",
"The CW"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6810,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1931
|
The Lazarus Effect is Donald Glover's highest-grossing film.
|
{
"evidence": [
"Donald Glover",
"Community (TV series)"
],
"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 Lazarus Effect is Donald Glover's highest-grossing film.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Donald Glover",
"Community (TV series)"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6811,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_635
|
Bus Stop is a 1956 American comedy-drama film directed by Joshua Logan that involves romantic love stories of the main characters about their journey through the period wherein a couple gets to know each other before a possible marriage.
|
{
"evidence": [
"Bus Stop (1956 film)",
"Romance film",
"Courtship"
],
"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:\nBus Stop is a 1956 American comedy-drama film directed by Joshua Logan that involves romantic love stories of the main characters about their journey through the period wherein a couple gets to know each other before a possible marriage.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Bus Stop (1956 film)",
"Romance film",
"Courtship"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6812,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
94475
|
In the Summer Games Tennis Tournament, Chung Hyeon ranked first in Group A while Gianluigi Quinzi ranked last in the group.
|
{
"evidence": [
"The Championships, Wimbledon_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:\nIn the Summer Games Tennis Tournament, Chung Hyeon ranked first in Group A while Gianluigi Quinzi ranked last in the group.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"The Championships, Wimbledon_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6813,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.402",
"value": "?search="
},
{
"operation": "search",
"time": "16.51",
"value": "Contains...Summer Games Tennis Tournament"
},
{
"operation": "Now on",
"time": "18.352",
"value": "undefined"
},
{
"operation": "search",
"time": "38.626",
"value": "Contains...Chung Hyeon ranked first in Group A"
},
{
"operation": "Now on",
"time": "41.49",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "46.566",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "47.616",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "53.041",
"value": "The Championships, Wimbledon"
},
{
"operation": "Now on",
"time": "55.546",
"value": "The Championships, Wimbledon"
},
{
"operation": "Highlighting",
"time": "63.56",
"value": "The Championships, Wimbledon_sentence_0"
},
{
"operation": "finish",
"time": "69.107",
"value": "finish"
}
]
}
|
exfever_2436
|
"I Believe" is a popular song written by an English songwriter.
|
{
"evidence": [
"I Believe (Frankie Laine song)",
"Ervin Drake"
],
"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:\n\"I Believe\" is a popular song written by an English songwriter.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"I Believe (Frankie Laine song)",
"Ervin Drake"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6814,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
86026
|
the 1964 Havering London Borough Council election was held on 7 May 1964 to elect members of Havering London Borough Council in London and was joined by a total of 146 candidates, Labour party(10), Conservative(44), Liberal (19) including 24 Independents and 4 Communists.
|
{
"evidence": [
"1964 Havering London Borough Council election_sentence_5",
"1964 Havering London Borough Council election_sentence_6",
"1964 Havering London Borough Council election_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:\nthe 1964 Havering London Borough Council election was held on 7 May 1964 to elect members of Havering London Borough Council in London and was joined by a total of 146 candidates, Labour party(10), Conservative(44), Liberal (19) including 24 Independents and 4 Communists.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"1964 Havering London Borough Council election_sentence_5",
"1964 Havering London Borough Council election_sentence_6",
"1964 Havering London Borough Council election_sentence_7"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6815,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.196",
"value": "?search="
},
{
"operation": "search",
"time": "11.854",
"value": "1964 Havering London Borough Council election"
},
{
"operation": "Now on",
"time": "16.669",
"value": "1964 Havering London Borough Council election"
},
{
"operation": "Highlighting",
"time": "20.603",
"value": "1964 Havering London Borough Council election_sentence_0"
},
{
"operation": "Highlighting",
"time": "32.195",
"value": "1964 Havering London Borough Council election_sentence_5"
},
{
"operation": "Highlighting",
"time": "54.95",
"value": "1964 Havering London Borough Council election_sentence_6"
},
{
"operation": "Highlighting",
"time": "59.766",
"value": "1964 Havering London Borough Council election_sentence_7"
},
{
"operation": "Highlighting deleted",
"time": "114.308",
"value": "1964 Havering London Borough Council election_sentence_0"
},
{
"operation": "finish",
"time": "147.929",
"value": "finish"
}
]
}
|
32855
|
Greg Ballard was one of many Republicans to file for mayor, but was greatly outspent by incumbent Peterson.
|
{
"evidence": [
"Greg Ballard_sentence_12"
],
"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:\nGreg Ballard was one of many Republicans to file for mayor, but was greatly outspent by incumbent Peterson.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Greg Ballard_sentence_12"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6816,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.947",
"value": "?search="
},
{
"operation": "search",
"time": "10.939",
"value": "Greg Ballard"
},
{
"operation": "Now on",
"time": "11.751",
"value": "Greg Ballard"
},
{
"operation": "Highlighting",
"time": "90.832",
"value": "Greg Ballard_sentence_12"
},
{
"operation": "finish",
"time": "101.597",
"value": "finish"
}
]
}
|
exfever_8858
|
Reds is a 1981 American historical drama film whose style of filmmaking is characterized by an extravagant setting and lavish costumes, accompanied by an expansive musical score.
|
{
"evidence": [
"Reds (film)",
"Epic 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:\nReds is a 1981 American historical drama film whose style of filmmaking is characterized by an extravagant setting and lavish costumes, accompanied by an expansive musical score.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Reds (film)",
"Epic film"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6817,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5492
|
The Fosters is a family drama series about a school vice-principal and a warranted law employee of a police force.
|
{
"evidence": [
"The Fosters (2013 TV series)",
"Police officer"
],
"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 Fosters is a family drama series about a school vice-principal and a warranted law employee of a police force.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Fosters (2013 TV series)",
"Police officer"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6818,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1015
|
EA Black Box was a video game developer based in Burnaby, British Columbia, America, primarily known for a racing video game franchise published by Electronic Arts and currently developed by Criterion Games, the developers of Burnout.
|
{
"evidence": [
"EA Black Box",
"Need for Speed"
],
"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:\nEA Black Box was a video game developer based in Burnaby, British Columbia, America, primarily known for a racing video game franchise published by Electronic Arts and currently developed by Criterion Games, the developers of Burnout.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"EA Black Box",
"Need for Speed"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6819,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7052
|
Spider-Man is an American comic book villain created by Stan Lee and an American comics artist and writer, best known as the artist and co-creator of the Marvel superhero Doctor Strange.
|
{
"evidence": [
"Spider-Man",
"Steve Ditko"
],
"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:\nSpider-Man is an American comic book villain created by Stan Lee and an American comics artist and writer, best known as the artist and co-creator of the Marvel superhero Doctor Strange.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Spider-Man",
"Steve Ditko"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6820,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8602
|
The Curious Case of Benjamin Button is the highest-grossing film of all the Brad Pitt films.
|
{
"evidence": [
"Brad Pitt",
"Legends of the Fall"
],
"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 Curious Case of Benjamin Button is the highest-grossing film of all the Brad Pitt films.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Brad Pitt",
"Legends of the Fall"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6821,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
52474
|
Jack Bird was born in Wollongong, and is of Indigenous American descent and played his junior rugby league for the Berkeley Eagles.
|
{
"evidence": [
"Jack Bird_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:\nJack Bird was born in Wollongong, and is of Indigenous American descent and played his junior rugby league for the Berkeley Eagles.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Jack Bird_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6822,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.9",
"value": "undefined"
},
{
"operation": "search",
"time": "8.847",
"value": "Jack Bird"
},
{
"operation": "Now on",
"time": "9.944",
"value": "Jack Bird"
},
{
"operation": "Highlighting",
"time": "18.58",
"value": "Jack Bird_sentence_4"
},
{
"operation": "finish",
"time": "26.128",
"value": "finish"
}
]
}
|
exfever_8842
|
Broadchurch is the most viewed crime drama series that has aired on ITV.
|
{
"evidence": [
"Broadchurch",
"Executive producer"
],
"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:\nBroadchurch is the most viewed crime drama series that has aired on ITV.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Broadchurch",
"Executive producer"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6823,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
77638
|
In 1922 banking operation and the ownership were always merged of Casse Toscane but the bank also formes a common holding company and one of the smallest share holders was Ente Cassa di Risparmio di Firenze.
|
{
"evidence": [
"Banca CR Firenze_sentence_5",
"Banca CR Firenze_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:\nIn 1922 banking operation and the ownership were always merged of Casse Toscane but the bank also formes a common holding company and one of the smallest share holders was Ente Cassa di Risparmio di Firenze.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Banca CR Firenze_sentence_5",
"Banca CR Firenze_sentence_7"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6824,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.865",
"value": "?search="
},
{
"operation": "search",
"time": "9.845",
"value": "1922 banking operation"
},
{
"operation": "Now on",
"time": "11.717",
"value": "undefined"
},
{
"operation": "search",
"time": "65.178",
"value": "casse toscane"
},
{
"operation": "Now on",
"time": "66.412",
"value": "Banca CR Firenze"
},
{
"operation": "Highlighting",
"time": "96.776",
"value": "Banca CR Firenze_sentence_5"
},
{
"operation": "Highlighting",
"time": "102.416",
"value": "Banca CR Firenze_sentence_7"
},
{
"operation": "finish",
"time": "107.768",
"value": "finish"
}
]
}
|
37166
|
The Wujiayuan Restaurant is one of only a few Chinese restaurants in Haiti, and it was arranged as a shelter during the 2010 Haiti earthquake.
|
{
"evidence": [
"Chinese Haitians_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:\nThe Wujiayuan Restaurant is one of only a few Chinese restaurants in Haiti, and it was arranged as a shelter during the 2010 Haiti earthquake.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Chinese Haitians_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6825,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.295",
"value": "?search="
},
{
"operation": "search",
"time": "7.818",
"value": "Wujiayuan Restaurant"
},
{
"operation": "Now on",
"time": "8.697",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "10.495",
"value": "Chinese Haitians"
},
{
"operation": "Now on",
"time": "11.537",
"value": "Chinese Haitians"
},
{
"operation": "Highlighting",
"time": "44.493",
"value": "Chinese Haitians_sentence_4"
},
{
"operation": "finish",
"time": "81.794",
"value": "finish"
}
]
}
|
49123
|
Melaleuca marginata is a prickly shrub with white flowers along the sides of the branches and tapered leaves.
|
{
"evidence": [
"Melaleuca marginata_sentence_3",
"Melaleuca marginata_sentence_4",
"Melaleuca marginata_sentence_5"
],
"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:\nMelaleuca marginata is a prickly shrub with white flowers along the sides of the branches and tapered leaves.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Melaleuca marginata_sentence_3",
"Melaleuca marginata_sentence_4",
"Melaleuca marginata_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6826,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.989",
"value": "?search="
},
{
"operation": "search",
"time": "17.114",
"value": "Melaleuca marginata"
},
{
"operation": "Now on",
"time": "17.843",
"value": "Melaleuca marginata"
},
{
"operation": "Highlighting",
"time": "21.997",
"value": "Melaleuca marginata_sentence_0"
},
{
"operation": "Highlighting",
"time": "27.036",
"value": "Melaleuca marginata_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "36.52",
"value": "Melaleuca marginata_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "37.499",
"value": "Melaleuca marginata_sentence_1"
},
{
"operation": "Highlighting",
"time": "39.265",
"value": "Melaleuca marginata_sentence_3"
},
{
"operation": "Highlighting",
"time": "43",
"value": "Melaleuca marginata_sentence_4"
},
{
"operation": "Highlighting",
"time": "50.504",
"value": "Melaleuca marginata_sentence_5"
},
{
"operation": "finish",
"time": "77.459",
"value": "finish"
}
]
}
|
exfever_104
|
Shraddha and Arjun Kapoor knew each other well as they belonged to the family of stars who both are known for sharing healthy friendships.
|
{
"evidence": [
"Half Girlfriend (film)",
"Shraddha Kapoor"
],
"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:\nShraddha and Arjun Kapoor knew each other well as they belonged to the family of stars who both are known for sharing healthy friendships.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Half Girlfriend (film)",
"Shraddha Kapoor"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6827,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
50973
|
Corel Designer, a vector-based graphics program, was originally developed by Micrografx, which was bought by Corel (a Canadian software company headquartered in Ottawa, Ontario, specializing in graphics processing) in 2001.
|
{
"evidence": [
"Corel Designer_sentence_0",
"Corel Designer_sentence_1",
"Corel_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:\nCorel Designer, a vector-based graphics program, was originally developed by Micrografx, which was bought by Corel (a Canadian software company headquartered in Ottawa, Ontario, specializing in graphics processing) in 2001.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Corel Designer_sentence_0",
"Corel Designer_sentence_1",
"Corel_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6828,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.931",
"value": "?search="
},
{
"operation": "search",
"time": "12.092",
"value": "Corel Designer"
},
{
"operation": "Now on",
"time": "13.49",
"value": "Corel Designer"
},
{
"operation": "Highlighting",
"time": "24.091",
"value": "Corel Designer_sentence_0"
},
{
"operation": "Highlighting",
"time": "27.817",
"value": "Corel Designer_sentence_1"
},
{
"operation": "hyperlink",
"time": "33.952",
"value": "Corel"
},
{
"operation": "Now on",
"time": "35.825",
"value": "Corel"
},
{
"operation": "Highlighting",
"time": "41.348",
"value": "Corel_sentence_1"
},
{
"operation": "finish",
"time": "49.849",
"value": "finish"
}
]
}
|
exfever_7670
|
Gone Girl is a 2014 American psychological horror film directed by David Fincher and written by an American writer.
|
{
"evidence": [
"Gone Girl (film)",
"Gillian Flynn"
],
"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:\nGone Girl is a 2014 American psychological horror film directed by David Fincher and written by an American writer.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Gone Girl (film)",
"Gillian Flynn"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6829,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1937
|
Paul Labile Pogba is a French professional baseballer for a?professional football club based in Old Trafford, Greater Manchester, England.
|
{
"evidence": [
"Paul Pogba",
"Manchester United F.C."
],
"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:\nPaul Labile Pogba is a French professional baseballer for a?professional football club based in Old Trafford, Greater Manchester, England.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Paul Pogba",
"Manchester United F.C."
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6830,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4982
|
Marie Antoinette Josèphe Jeanne was the penultimate child and youngest daughter of the ruler of the Habsburg dominions from 1740 until her death in 1780.
|
{
"evidence": [
"Marie Antoinette",
"Maria Theresa"
],
"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:\nMarie Antoinette Josèphe Jeanne was the penultimate child and youngest daughter of the ruler of the Habsburg dominions from 1740 until her death in 1780.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Marie Antoinette",
"Maria Theresa"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6831,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3509
|
Tropic Thunder is a 2008 film starring an American actor and stand-up comedian best known for his role in the film Roll Bounce in 2005.
|
{
"evidence": [
"Tropic Thunder",
"Brandon T. Jackson"
],
"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:\nTropic Thunder is a 2008 film starring an American actor and stand-up comedian best known for his role in the film Roll Bounce in 2005.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tropic Thunder",
"Brandon T. Jackson"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6832,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6484
|
Edward Harrison Norton is an Italian actor and filmmaker who graduated from the undergraduate college of Yale University, which was founded in 1701.
|
{
"evidence": [
"Edward Norton",
"Yale College"
],
"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:\nEdward Harrison Norton is an Italian actor and filmmaker who graduated from the undergraduate college of Yale University, which was founded in 1701.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Edward Norton",
"Yale College"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6833,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
74055
|
Joe Magidsohn scored one touchdown with a total of ten points, while Frederick L. Conklin scored three extra points and one field goal with a total of six points.
|
{
"evidence": [
"Joe Magidsohn_sentence_0",
"Joe Magidsohn_sentence_1"
],
"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:\nJoe Magidsohn scored one touchdown with a total of ten points, while Frederick L. Conklin scored three extra points and one field goal with a total of six points.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Joe Magidsohn_sentence_0",
"Joe Magidsohn_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6834,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.099",
"value": "?search="
},
{
"operation": "search",
"time": "12.901",
"value": "Joe Magidsohn"
},
{
"operation": "Now on",
"time": "13.793",
"value": "Joe Magidsohn"
},
{
"operation": "Highlighting",
"time": "158.996",
"value": "Joe Magidsohn_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "160.66",
"value": "Joe Magidsohn_sentence_0"
},
{
"operation": "Page search",
"time": "184.789",
"value": "frederick"
},
{
"operation": "hyperlink",
"time": "202.508",
"value": "University of Michigan"
},
{
"operation": "Now on",
"time": "207.533",
"value": "University of Michigan"
},
{
"operation": "search",
"time": "230.235",
"value": "1909 University of Michigan football"
},
{
"operation": "Now on",
"time": "233.412",
"value": "undefined"
},
{
"operation": "search",
"time": "249.311",
"value": "1909 college football season"
},
{
"operation": "Now on",
"time": "252.401",
"value": "1909 college football season"
},
{
"operation": "Page search",
"time": "261.784",
"value": "magidoshn"
},
{
"operation": "search",
"time": "279.274",
"value": "frederick l conklin"
},
{
"operation": "Now on",
"time": "280.64",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "290.533",
"value": "1911 Michigan Wolverines football team"
},
{
"operation": "Now on",
"time": "291.478",
"value": "1911 Michigan Wolverines football team"
},
{
"operation": "Page search",
"time": "388.68",
"value": "magidshon"
},
{
"operation": "Page search",
"time": "398.358",
"value": "magidsohn"
},
{
"operation": "search",
"time": "436.592",
"value": "Joe Magidsohn"
},
{
"operation": "Now on",
"time": "437.397",
"value": "Joe Magidsohn"
},
{
"operation": "Highlighting",
"time": "464.189",
"value": "Joe Magidsohn_sentence_0"
},
{
"operation": "Highlighting",
"time": "466.996",
"value": "Joe Magidsohn_sentence_1"
},
{
"operation": "Page search",
"time": "476.5",
"value": "field goal"
},
{
"operation": "Page search",
"time": "482.909",
"value": "touchdown"
},
{
"operation": "finish",
"time": "506.26",
"value": "finish"
}
]
}
|
exfever_1890
|
To Die For is a 1996 film directed by Gus Van Sant that starred an American actor who has received various accolades, including an Oscar and Grammy nomination.
|
{
"evidence": [
"To Die For",
"Matt Dillon"
],
"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:\nTo Die For is a 1996 film directed by Gus Van Sant that starred an American actor who has received various accolades, including an Oscar and Grammy nomination.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"To Die For",
"Matt Dillon"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6835,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
40072
|
Ralph Lewis was the youngest of seven son, hosted TV shows and moved to Niagara Falls which has a population of 50,193.
|
{
"evidence": [
"Sons of Ralph_sentence_6"
],
"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:\nRalph Lewis was the youngest of seven son, hosted TV shows and moved to Niagara Falls which has a population of 50,193.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Sons of Ralph_sentence_6"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6836,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.755",
"value": "?search="
},
{
"operation": "search",
"time": "10.57",
"value": "Ralph Lewis"
},
{
"operation": "Now on",
"time": "11.168",
"value": "Ralph Lewis"
},
{
"operation": "hyperlink",
"time": "25.331",
"value": "Ralph Lewis (basketball)"
},
{
"operation": "Now on",
"time": "26.073",
"value": "Ralph Lewis (basketball)"
},
{
"operation": "back-button-clicked",
"time": "34.838",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "34.925",
"value": "Ralph Lewis"
},
{
"operation": "hyperlink",
"time": "36.466",
"value": "Ralph Lewis (actor)"
},
{
"operation": "Now on",
"time": "36.979",
"value": "Ralph Lewis (actor)"
},
{
"operation": "back-button-clicked",
"time": "54.268",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "54.359",
"value": "Ralph Lewis"
},
{
"operation": "search",
"time": "73.857",
"value": "youngest of seven sons"
},
{
"operation": "Now on",
"time": "76.671",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "123.591",
"value": "Sons of Ralph"
},
{
"operation": "Now on",
"time": "124.321",
"value": "Sons of Ralph"
},
{
"operation": "Highlighting",
"time": "142.486",
"value": "Sons of Ralph_sentence_5"
},
{
"operation": "Highlighting",
"time": "148.91",
"value": "Sons of Ralph_sentence_7"
},
{
"operation": "Highlighting",
"time": "172.845",
"value": "Sons of Ralph_sentence_6"
},
{
"operation": "Highlighting deleted",
"time": "174.477",
"value": "Sons of Ralph_sentence_5"
},
{
"operation": "Highlighting deleted",
"time": "175.396",
"value": "Sons of Ralph_sentence_7"
},
{
"operation": "finish",
"time": "185.253",
"value": "finish"
}
]
}
|
66150
|
Russian, wild forest reindeer, similar to North American woodland caribou, reached their peak population in the 1800's.
|
{
"evidence": [
"Reindeer in Russia_sentence_40",
"Reindeer in Russia_sentence_5"
],
"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:\nRussian, wild forest reindeer, similar to North American woodland caribou, reached their peak population in the 1800's.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Reindeer in Russia_sentence_40",
"Reindeer in Russia_sentence_5"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6837,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.142",
"value": "?search="
},
{
"operation": "search",
"time": "28.986",
"value": "Contains...Russian wild forest reindeer,"
},
{
"operation": "Now on",
"time": "30.726",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "46.385",
"value": "Reindeer"
},
{
"operation": "Now on",
"time": "47.588",
"value": "Reindeer"
},
{
"operation": "Page search",
"time": "53.698",
"value": "russian"
},
{
"operation": "hyperlink",
"time": "68.341",
"value": "Russian"
},
{
"operation": "Now on",
"time": "71.915",
"value": "Reindeer"
},
{
"operation": "back-button-clicked",
"time": "111.14",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "111.156",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "111.875",
"value": "undefined"
},
{
"operation": "search",
"time": "128.109",
"value": "Contains...Russian wild forest reindeer"
},
{
"operation": "Now on",
"time": "129.546",
"value": "undefined"
},
{
"operation": "search",
"time": "166.28",
"value": "Contains...North American woodland caribou"
},
{
"operation": "Now on",
"time": "169.09",
"value": "undefined"
},
{
"operation": "search",
"time": "186.471",
"value": "Contains...Russian, wild forest reindeer, similar to North American woodland caribou, reached their peak population in the 1800's."
},
{
"operation": "Now on",
"time": "188.701",
"value": "undefined"
},
{
"operation": "search",
"time": "207.529",
"value": "Contains...wild forest reindeer"
},
{
"operation": "Now on",
"time": "208.504",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "220.604",
"value": "Reindeer in Russia"
},
{
"operation": "Now on",
"time": "221.215",
"value": "Reindeer in Russia"
},
{
"operation": "Page search",
"time": "229.927",
"value": "woodland"
},
{
"operation": "Highlighting",
"time": "239.649",
"value": "Reindeer in Russia_sentence_40"
},
{
"operation": "Page search",
"time": "258.664",
"value": "population"
},
{
"operation": "Page search",
"time": "276.548",
"value": "peak population"
},
{
"operation": "Page search",
"time": "288.875",
"value": "population"
},
{
"operation": "Highlighting",
"time": "316.112",
"value": "Reindeer in Russia_sentence_47"
},
{
"operation": "Highlighting deleted",
"time": "320.584",
"value": "Reindeer in Russia_sentence_47"
},
{
"operation": "Highlighting",
"time": "322.04",
"value": "Reindeer in Russia_sentence_52"
},
{
"operation": "Highlighting deleted",
"time": "326.266",
"value": "Reindeer in Russia_sentence_52"
},
{
"operation": "Page search",
"time": "343.546",
"value": "highest population"
},
{
"operation": "Page search",
"time": "361.824",
"value": "population"
},
{
"operation": "Highlighting",
"time": "380.934",
"value": "Reindeer in Russia_sentence_16"
},
{
"operation": "Highlighting",
"time": "380.935",
"value": "Reindeer in Russia_section_1"
},
{
"operation": "Highlighting",
"time": "380.935",
"value": "Reindeer in Russia_section_2"
},
{
"operation": "Highlighting deleted",
"time": "384.18",
"value": "Reindeer in Russia_sentence_16"
},
{
"operation": "Highlighting deleted",
"time": "384.183",
"value": "Reindeer in Russia_section_1"
},
{
"operation": "Highlighting deleted",
"time": "384.185",
"value": "Reindeer in Russia_section_2"
},
{
"operation": "Highlighting",
"time": "391.109",
"value": "Reindeer in Russia_sentence_26"
},
{
"operation": "Highlighting",
"time": "391.11",
"value": "Reindeer in Russia_section_3"
},
{
"operation": "Highlighting deleted",
"time": "403.184",
"value": "Reindeer in Russia_sentence_26"
},
{
"operation": "Highlighting deleted",
"time": "403.185",
"value": "Reindeer in Russia_section_3"
},
{
"operation": "Highlighting",
"time": "409.447",
"value": "Reindeer in Russia_sentence_47"
},
{
"operation": "Highlighting deleted",
"time": "412.796",
"value": "Reindeer in Russia_sentence_47"
},
{
"operation": "Highlighting",
"time": "414.57",
"value": "Reindeer in Russia_sentence_52"
},
{
"operation": "Highlighting deleted",
"time": "428.963",
"value": "Reindeer in Russia_sentence_52"
},
{
"operation": "Highlighting",
"time": "430.422",
"value": "Reindeer in Russia_sentence_53"
},
{
"operation": "Highlighting",
"time": "430.423",
"value": "Reindeer in Russia_section_6"
},
{
"operation": "Highlighting deleted",
"time": "435.945",
"value": "Reindeer in Russia_sentence_53"
},
{
"operation": "Highlighting deleted",
"time": "435.946",
"value": "Reindeer in Russia_section_6"
},
{
"operation": "Page search",
"time": "456.712",
"value": "1800"
},
{
"operation": "Page search",
"time": "463.533",
"value": "18"
},
{
"operation": "Highlighting",
"time": "469.727",
"value": "Reindeer in Russia_sentence_5"
},
{
"operation": "Highlighting",
"time": "485.729",
"value": "Reindeer in Russia_sentence_18"
},
{
"operation": "Highlighting",
"time": "485.73",
"value": "Reindeer in Russia_section_1"
},
{
"operation": "Highlighting",
"time": "485.73",
"value": "Reindeer in Russia_section_2"
},
{
"operation": "Highlighting deleted",
"time": "500.2",
"value": "Reindeer in Russia_sentence_18"
},
{
"operation": "Highlighting deleted",
"time": "500.202",
"value": "Reindeer in Russia_section_1"
},
{
"operation": "Highlighting deleted",
"time": "500.203",
"value": "Reindeer in Russia_section_2"
},
{
"operation": "Highlighting",
"time": "508.138",
"value": "Reindeer in Russia_sentence_67"
},
{
"operation": "Highlighting",
"time": "508.139",
"value": "Reindeer in Russia_section_8"
},
{
"operation": "Highlighting deleted",
"time": "513.184",
"value": "Reindeer in Russia_sentence_67"
},
{
"operation": "Highlighting deleted",
"time": "513.185",
"value": "Reindeer in Russia_section_8"
},
{
"operation": "finish",
"time": "527.808",
"value": "finish"
}
]
}
|
exfever_5716
|
Tamera Darvette Mowry rose to prominence as Tamera Campbell, a teen on the sitcom Sister, Sister, alongside her twin who also rose to fame in the same series.
|
{
"evidence": [
"Tamera Mowry",
"Tia Mowry"
],
"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:\nTamera Darvette Mowry rose to prominence as Tamera Campbell, a teen on the sitcom Sister, Sister, alongside her twin who also rose to fame in the same series.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tamera Mowry",
"Tia Mowry"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6838,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
42549
|
Angela Black was a Vanderbilt University graduate who worked in KABC-TV as a reporter and weekend anchor for twelve years.
|
{
"evidence": [
"Angela Black_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:\nAngela Black was a Vanderbilt University graduate who worked in KABC-TV as a reporter and weekend anchor for twelve years.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Angela Black_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6839,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.707",
"value": "?search="
},
{
"operation": "search",
"time": "3.708",
"value": ""
},
{
"operation": "Now on",
"time": "4.091",
"value": "undefined"
},
{
"operation": "search",
"time": "9.341",
"value": "angela Black"
},
{
"operation": "Now on",
"time": "9.903",
"value": "Angela Black"
},
{
"operation": "Highlighting",
"time": "24.383",
"value": "Angela Black_sentence_2"
},
{
"operation": "finish",
"time": "38.378",
"value": "finish"
}
]
}
|
exfever_5958
|
Heather Joan Graham is an American actress who made her first starring role in the 1988 American teen comedy film written by a Canadian screenwriter and film director from Montreal.
|
{
"evidence": [
"Heather Graham",
"License to Drive",
"Neil Tolkin"
],
"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:\nHeather Joan Graham is an American actress who made her first starring role in the 1988 American teen comedy film written by a Canadian screenwriter and film director from Montreal.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Heather Graham",
"License to Drive",
"Neil Tolkin"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6840,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2328
|
Terminator Salvation is an American military science fiction action film starring an American actor who appeared in a comedy-drama on an American premium television network owned by Paramount Media Networks.
|
{
"evidence": [
"Terminator Salvation",
"Anton Yelchin",
"Showtime (TV network)"
],
"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:\nTerminator Salvation is an American military science fiction action film starring an American actor who appeared in a comedy-drama on an American premium television network owned by Paramount Media Networks.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Terminator Salvation",
"Anton Yelchin",
"Showtime (TV network)"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6841,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5940
|
The Clocks Higher Education World University Rankings is a yearly publication by Quacquarelli Symonds (QS).
|
{
"evidence": [
"Times Higher Education World University Rankings",
"QS World University Rankings"
],
"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 Clocks Higher Education World University Rankings is a yearly publication by Quacquarelli Symonds (QS).\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Times Higher Education World University Rankings",
"QS World University Rankings"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6842,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
89871
|
At Barack Obama Tucson memorial speech he mentioned visiting Gabby and talked about how people wanted to make sense of things by debating.
|
{
"evidence": [
"Barack Obama Tucson memorial speech_sentence_20",
"Barack Obama Tucson memorial speech_sentence_21"
],
"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:\nAt Barack Obama Tucson memorial speech he mentioned visiting Gabby and talked about how people wanted to make sense of things by debating.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Barack Obama Tucson memorial speech_sentence_20",
"Barack Obama Tucson memorial speech_sentence_21"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6843,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.785",
"value": "?search="
},
{
"operation": "search",
"time": "24.311",
"value": "barack obama tuscon memorial"
},
{
"operation": "Now on",
"time": "25.283",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "30.663",
"value": "2011 Tucson shooting"
},
{
"operation": "Now on",
"time": "31.34",
"value": "2011 Tucson shooting"
},
{
"operation": "Page search",
"time": "50.953",
"value": "obama"
},
{
"operation": "Highlighting",
"time": "71.616",
"value": "2011 Tucson shooting_sentence_114"
},
{
"operation": "Highlighting",
"time": "71.617",
"value": "2011 Tucson shooting_title"
},
{
"operation": "Highlighting",
"time": "71.617",
"value": "2011 Tucson shooting_section_4"
},
{
"operation": "Highlighting",
"time": "71.617",
"value": "2011 Tucson shooting_section_5"
},
{
"operation": "Highlighting deleted",
"time": "81.543",
"value": "2011 Tucson shooting_sentence_114"
},
{
"operation": "Highlighting deleted",
"time": "81.544",
"value": "2011 Tucson shooting_title"
},
{
"operation": "Highlighting deleted",
"time": "81.545",
"value": "2011 Tucson shooting_section_4"
},
{
"operation": "Highlighting deleted",
"time": "81.545",
"value": "2011 Tucson shooting_section_5"
},
{
"operation": "Highlighting",
"time": "95.439",
"value": "2011 Tucson shooting_sentence_159"
},
{
"operation": "Highlighting",
"time": "95.44",
"value": "2011 Tucson shooting_title"
},
{
"operation": "Highlighting",
"time": "95.44",
"value": "2011 Tucson shooting_section_6"
},
{
"operation": "Highlighting",
"time": "95.44",
"value": "2011 Tucson shooting_section_9"
},
{
"operation": "hyperlink",
"time": "102.016",
"value": "a memorial speech"
},
{
"operation": "Now on",
"time": "102.73",
"value": "Barack Obama Tucson memorial speech"
},
{
"operation": "Page search",
"time": "127.303",
"value": "gabby"
},
{
"operation": "Highlighting",
"time": "134.006",
"value": "Barack Obama Tucson memorial speech_sentence_20"
},
{
"operation": "Highlighting",
"time": "161.318",
"value": "Barack Obama Tucson memorial speech_sentence_18"
},
{
"operation": "Highlighting",
"time": "175.822",
"value": "Barack Obama Tucson memorial speech_sentence_21"
},
{
"operation": "Highlighting deleted",
"time": "218.799",
"value": "Barack Obama Tucson memorial speech_sentence_18"
},
{
"operation": "finish",
"time": "224.158",
"value": "finish"
}
]
}
|
6681
|
The Maracot Deep is a short 1929 poem by Arthur Conan Doyle about the discovery of a sunken city of Atlantis, an ancient city or continent which was drowned by the sea due to divine intervention, its discovery is by a team of explorers led by Professor Maracot.
|
{
"evidence": [
"The Maracot Deep_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:\nThe Maracot Deep is a short 1929 poem by Arthur Conan Doyle about the discovery of a sunken city of Atlantis, an ancient city or continent which was drowned by the sea due to divine intervention, its discovery is by a team of explorers led by Professor Maracot.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"The Maracot Deep_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6844,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.419",
"value": "?search="
},
{
"operation": "search",
"time": "7.672",
"value": "The Maracot Deep"
},
{
"operation": "Now on",
"time": "8.716",
"value": "The Maracot Deep"
},
{
"operation": "Highlighting",
"time": "9.995",
"value": "The Maracot Deep_sentence_0"
},
{
"operation": "finish",
"time": "20.668",
"value": "finish"
}
]
}
|
exfever_3255
|
The Silence of the Lambs is a 1992 American psychological horror film directed by Jonathan Demme and written by an American screenwriter and a person who writes plays.
|
{
"evidence": [
"The Silence of the Lambs (film)",
"Ted Tally",
"Playwright"
],
"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 Silence of the Lambs is a 1992 American psychological horror film directed by Jonathan Demme and written by an American screenwriter and a person who writes plays.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Silence of the Lambs (film)",
"Ted Tally",
"Playwright"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6845,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2594
|
Macbeth is a love story by William Shakespeare, first published in the Comedies, Histories, & Tragedies, a collection of plays.
|
{
"evidence": [
"Macbeth",
"First Folio"
],
"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:\nMacbeth is a love story by William Shakespeare, first published in the Comedies, Histories, & Tragedies, a collection of plays.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Macbeth",
"First Folio"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6846,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
59162
|
Lord Vishnu changed his appearance to be able to follow Aryamala.
|
{
"evidence": [
"Aryamala_sentence_19",
"Aryamala_sentence_23"
],
"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:\nLord Vishnu changed his appearance to be able to follow Aryamala.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Aryamala_sentence_19",
"Aryamala_sentence_23"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6847,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.959",
"value": "?search="
},
{
"operation": "search",
"time": "15.892",
"value": "Lord Vishnu"
},
{
"operation": "Now on",
"time": "19.046",
"value": "Vishnu"
},
{
"operation": "search",
"time": "83.607",
"value": "Lord Vishnu Temples in Kerala"
},
{
"operation": "Now on",
"time": "84.916",
"value": "List of Hindu temples in Kerala"
},
{
"operation": "search",
"time": "97.096",
"value": "Lord Vishnu aryamala"
},
{
"operation": "Now on",
"time": "97.736",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "99.99",
"value": "Aryamala"
},
{
"operation": "Now on",
"time": "100.558",
"value": "Aryamala"
},
{
"operation": "search",
"time": "162.237",
"value": "Lord Vishnu"
},
{
"operation": "Now on",
"time": "162.838",
"value": "Lord Vishnu"
},
{
"operation": "search",
"time": "230.6",
"value": "Aryamala"
},
{
"operation": "Now on",
"time": "231.065",
"value": "Aryamala"
},
{
"operation": "Highlighting",
"time": "253.126",
"value": "Aryamala_sentence_19"
},
{
"operation": "Highlighting",
"time": "258.277",
"value": "Aryamala_sentence_23"
},
{
"operation": "search",
"time": "355.191",
"value": "Prahalada (film)"
},
{
"operation": "Now on",
"time": "356.393",
"value": "Prahalada (film)"
},
{
"operation": "finish",
"time": "376.83",
"value": "finish"
}
]
}
|
exfever_1374
|
Harold Ramis was an American writer who directed a sports comedy film written by Brian Doyle-Murray.
|
{
"evidence": [
"Harold Ramis",
"Caddyshack"
],
"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:\nHarold Ramis was an American writer who directed a sports comedy film written by Brian Doyle-Murray.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Harold Ramis",
"Caddyshack"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6848,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2037
|
Sigmund Freud was an Austrian neurologist and the founder of psychoanalysis, born in Central-Eastern European multinational great power from 1804 to 1867.
|
{
"evidence": [
"Sigmund Freud",
"Austrian Empire"
],
"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:\nSigmund Freud was an Austrian neurologist and the founder of psychoanalysis, born in Central-Eastern European multinational great power from 1804 to 1867.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Sigmund Freud",
"Austrian Empire"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6849,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4818
|
The District of North Vancouver is a designation for a class of municipal in Canada.
|
{
"evidence": [
"North Vancouver (district municipality)",
"District municipality"
],
"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 District of North Vancouver is a designation for a class of municipal in Canada.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"North Vancouver (district municipality)",
"District municipality"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6850,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
24298
|
Ibrahim Hassan, an Egyptian former professional football player who played as a striker, managed a total of 12 Egyptian teams and 1 Jordan team from 2008 to 2020.
|
{
"evidence": [
"Ibrahim Hassan_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:\nIbrahim Hassan, an Egyptian former professional football player who played as a striker, managed a total of 12 Egyptian teams and 1 Jordan team from 2008 to 2020.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Ibrahim Hassan_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6851,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.662",
"value": "?search="
},
{
"operation": "search",
"time": "14.37",
"value": "Ibrahim Hassan"
},
{
"operation": "Now on",
"time": "15.581",
"value": "Ibrahim Hassan"
},
{
"operation": "Highlighting",
"time": "20.994",
"value": "Ibrahim Hassan_sentence_2"
},
{
"operation": "finish",
"time": "33.747",
"value": "finish"
}
]
}
|
exfever_1963
|
The Apple Store is a chain of retail stores owned and operated by a British multinational technology company that specializes in consumer electronics, software and online services.
|
{
"evidence": [
"Apple Store",
"Apple Inc."
],
"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 Apple Store is a chain of retail stores owned and operated by a British multinational technology company that specializes in consumer electronics, software and online services.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Apple Store",
"Apple Inc."
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6852,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3615
|
Midtjylland was the most populous football club in Denmark.
|
{
"evidence": [
"Marcus Rashford",
"UEFA Europa League"
],
"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:\nMidtjylland was the most populous football club in Denmark.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Marcus Rashford",
"UEFA Europa League"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6853,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
3052
|
Helge von Koch was an Australian painter who was born to Swedish nobility and received his Ph.D. in 1892.
|
{
"evidence": [
"Helge von Koch_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:\nHelge von Koch was an Australian painter who was born to Swedish nobility and received his Ph.D. in 1892.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Helge von Koch_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6854,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "search",
"time": "12.614",
"value": "Helge von Koch"
},
{
"operation": "search",
"time": "47.058",
"value": "Contains...Helge von Koch"
},
{
"operation": "back-button-clicked",
"time": "60.105",
"value": "back-button-clicked"
},
{
"operation": "Highlighting",
"time": "63.158",
"value": "Helge von Koch_sentence_0"
},
{
"operation": "finish",
"time": "88.024",
"value": "finish"
}
]
}
|
exfever_4668
|
Joe Biden is also a member of the Democratic Party.
|
{
"evidence": [
"John McCain",
"Barack Obama",
"Columbia University"
],
"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:\nJoe Biden is also a member of the Democratic Party.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"John McCain",
"Barack Obama",
"Columbia University"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6855,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_557
|
Nikolaj William Coster-Waldau is a Danish actor who has appeared in numerous films, including a 2011 Norwegian action thriller film based on the 2008 novel of the same name by Mark Wahlberg.
|
{
"evidence": [
"Nikolaj Coster-Waldau",
"Headhunters (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:\nNikolaj William Coster-Waldau is a Danish actor who has appeared in numerous films, including a 2011 Norwegian action thriller film based on the 2008 novel of the same name by Mark Wahlberg.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Nikolaj Coster-Waldau",
"Headhunters (film)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6856,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
80606
|
FK Naša Krila Zemun was formed by the French after World War I to compete in the Yugoslav League system where they would finish fifth in the national top league.
|
{
"evidence": [
"FK Naša Krila Zemun_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:\nFK Naša Krila Zemun was formed by the French after World War I to compete in the Yugoslav League system where they would finish fifth in the national top league.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"FK Naša Krila Zemun_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6857,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.874",
"value": "?search="
},
{
"operation": "search",
"time": "45.981",
"value": "FK Naša Krila Zemun"
},
{
"operation": "Now on",
"time": "47.729",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "52.559",
"value": "FK Naša Krila Zemun"
},
{
"operation": "Now on",
"time": "53.12",
"value": "FK Naša Krila Zemun"
},
{
"operation": "Highlighting",
"time": "61.478",
"value": "FK Naša Krila Zemun_sentence_2"
},
{
"operation": "finish",
"time": "138.399",
"value": "finish"
}
]
}
|
exfever_2503
|
Andrew Jackson served briefly in the United States House of Representatives and the Senate, representing a state in the Southeastern region that is bordered to the east by the 9th-most populous state in the U.S.
|
{
"evidence": [
"Andrew Jackson",
"Tennessee",
"North Carolina"
],
"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:\nAndrew Jackson served briefly in the United States House of Representatives and the Senate, representing a state in the Southeastern region that is bordered to the east by the 9th-most populous state in the U.S.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Andrew Jackson",
"Tennessee",
"North Carolina"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6858,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
33754
|
The source of the water for the fountain of Piazza Colonna was only 67 feet above sea level; like the Trevi Fountain (designed by Italian architect Nicola Salvi), the Fontana della Barcaccia, and the fountains of Piazza Navona, all connected to the Acqua Vergine, the fountain of Piazza Colonna, built by Della Porta, was unable to jet water high into the air.
|
{
"evidence": [
"Fontana di Piazza Colonna_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:\nThe source of the water for the fountain of Piazza Colonna was only 67 feet above sea level; like the Trevi Fountain (designed by Italian architect Nicola Salvi), the Fontana della Barcaccia, and the fountains of Piazza Navona, all connected to the Acqua Vergine, the fountain of Piazza Colonna, built by Della Porta, was unable to jet water high into the air.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Fontana di Piazza Colonna_sentence_6"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6859,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.687",
"value": "?search="
},
{
"operation": "search",
"time": "19.487",
"value": "Contains...fountain of Piazza Colonna"
},
{
"operation": "Now on",
"time": "20.682",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "21.683",
"value": "Fontana di Piazza Colonna"
},
{
"operation": "Now on",
"time": "22.351",
"value": "Fontana di Piazza Colonna"
},
{
"operation": "Highlighting",
"time": "24.59",
"value": "Fontana di Piazza Colonna_sentence_0"
},
{
"operation": "Highlighting",
"time": "31.835",
"value": "Fontana di Piazza Colonna_sentence_6"
},
{
"operation": "Highlighting deleted",
"time": "33.281",
"value": "Fontana di Piazza Colonna_sentence_0"
},
{
"operation": "finish",
"time": "329.809",
"value": "finish"
}
]
}
|
exfever_1395
|
War of the Worlds is a 2005 American science fiction action film based on the 1898 novel by an Irish writer who wrote dozens of novels, short stories, and works of social commentary, history, satire, biography, and autobiography.
|
{
"evidence": [
"War of the Worlds (2005 film)",
"H. G. Wells"
],
"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:\nWar of the Worlds is a 2005 American science fiction action film based on the 1898 novel by an Irish writer who wrote dozens of novels, short stories, and works of social commentary, history, satire, biography, and autobiography.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"War of the Worlds (2005 film)",
"H. G. Wells"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6860,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
58711
|
Gravelbourg, located in the Gravelbourg No. 104, Saskatchewan, neither a prairie nor a boreal province in Canada, has a total population of 1,116 and covers an area of 3.23 km (1.25 sq mi).
|
{
"evidence": [
"Saskatchewan_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:\nGravelbourg, located in the Gravelbourg No. 104, Saskatchewan, neither a prairie nor a boreal province in Canada, has a total population of 1,116 and covers an area of 3.23 km (1.25 sq mi).\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Saskatchewan_sentence_3"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6861,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.384",
"value": "?search="
},
{
"operation": "search",
"time": "9.062",
"value": "Gravelbourg"
},
{
"operation": "Now on",
"time": "9.95",
"value": "Gravelbourg"
},
{
"operation": "Highlighting",
"time": "18.078",
"value": "Gravelbourg_header_cell_0_3_0"
},
{
"operation": "Highlighting",
"time": "18.08",
"value": "Gravelbourg_title"
},
{
"operation": "Highlighting",
"time": "19.668",
"value": "Gravelbourg_cell_0_3_1"
},
{
"operation": "Highlighting",
"time": "19.687",
"value": "Gravelbourg_header_cell_0_3_0"
},
{
"operation": "Highlighting",
"time": "19.687",
"value": "Gravelbourg_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "20.854",
"value": "Gravelbourg_cell_0_2_1"
},
{
"operation": "Highlighting",
"time": "20.869",
"value": "Gravelbourg_header_cell_0_2_0"
},
{
"operation": "Highlighting",
"time": "42.468",
"value": "Gravelbourg_cell_0_1_1"
},
{
"operation": "Highlighting",
"time": "42.483",
"value": "Gravelbourg_header_cell_0_1_0"
},
{
"operation": "hyperlink",
"time": "43.386",
"value": "Saskatchewan"
},
{
"operation": "Now on",
"time": "50.019",
"value": "Saskatchewan"
},
{
"operation": "Highlighting",
"time": "51.391",
"value": "Saskatchewan_sentence_3"
},
{
"operation": "finish",
"time": "64.317",
"value": "finish"
}
]
}
|
exfever_4921
|
Broadchurch is a British crime drama television series broadcast on a British free-to-air television network, known as the oldest commercial network in the UK.
|
{
"evidence": [
"Broadchurch",
"ITV (TV network)"
],
"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:\nBroadchurch is a British crime drama television series broadcast on a British free-to-air television network, known as the oldest commercial network in the UK.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Broadchurch",
"ITV (TV network)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6862,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7082
|
Elsa Lafuente Medianu, known professionally as Elsa Pataky, is a Spanish model and actress who appeared in the 2009 American neo-noir crime thriller movie directed by Russell Mulcahy.
|
{
"evidence": [
"Elsa Pataky",
"Give 'Em Hell, Malone"
],
"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:\nElsa Lafuente Medianu, known professionally as Elsa Pataky, is a Spanish model and actress who appeared in the 2009 American neo-noir crime thriller movie directed by Russell Mulcahy.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Elsa Pataky",
"Give 'Em Hell, Malone"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6863,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4824
|
Sammy Cahn was the first musician to collaborate with Frank Sinatra.
|
{
"evidence": [
"Sammy Cahn",
"Doris Day"
],
"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:\nSammy Cahn was the first musician to collaborate with Frank Sinatra.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Sammy Cahn",
"Doris Day"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6864,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
15014
|
The Danish footballer for Boldklubben 1908 who used to play with two clubs from 2011 to 2013 appearing in 47 games making 16 goals in total is Nichola Rohde.
|
{
"evidence": [
"Boldklubben 1908_sentence_0",
"Boldklubben 1908_sentence_1",
"Boldklubben 1908_sentence_4",
"Boldklubben 1908_sentence_5",
"Boldklubben 1908_sentence_6",
"List of Boldklubben 1908 seasons_sentence_0",
"List of Boldklubben 1908 seasons_sentence_5"
],
"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 Danish footballer for Boldklubben 1908 who used to play with two clubs from 2011 to 2013 appearing in 47 games making 16 goals in total is Nichola Rohde.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Boldklubben 1908_sentence_0",
"Boldklubben 1908_sentence_1",
"Boldklubben 1908_sentence_4",
"Boldklubben 1908_sentence_5",
"Boldklubben 1908_sentence_6",
"List of Boldklubben 1908 seasons_sentence_0",
"List of Boldklubben 1908 seasons_sentence_5"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6865,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.887",
"value": "?search="
},
{
"operation": "search",
"time": "30.511",
"value": "nichola rhode"
},
{
"operation": "Now on",
"time": "32.008",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "34.131",
"value": "nichols rhode"
},
{
"operation": "Now on",
"time": "35.77",
"value": "undefined"
},
{
"operation": "search",
"time": "50.497",
"value": "Boldklubben 1909"
},
{
"operation": "Now on",
"time": "52.025",
"value": "Boldklubben 1909"
},
{
"operation": "search",
"time": "110.864",
"value": "nichol* rohde"
},
{
"operation": "Now on",
"time": "112.706",
"value": "undefined"
},
{
"operation": "search",
"time": "121.735",
"value": "nichol rohde"
},
{
"operation": "Now on",
"time": "122.788",
"value": "undefined"
},
{
"operation": "search",
"time": "145.496",
"value": "List of Boldklubben 1908 seasons"
},
{
"operation": "Now on",
"time": "147.038",
"value": "List of Boldklubben 1908 seasons"
},
{
"operation": "search",
"time": "164.978",
"value": "Boldklubben 1908"
},
{
"operation": "Now on",
"time": "166.805",
"value": "Boldklubben 1908"
},
{
"operation": "search",
"time": "194.601",
"value": "List of Boldklubben 1908 seasons"
},
{
"operation": "Now on",
"time": "195.752",
"value": "List of Boldklubben 1908 seasons"
},
{
"operation": "hyperlink",
"time": "204.47",
"value": "B 1908"
},
{
"operation": "Now on",
"time": "205.363",
"value": "Boldklubben 1908"
},
{
"operation": "Highlighting",
"time": "223.27",
"value": "Boldklubben 1908_sentence_0"
},
{
"operation": "Highlighting",
"time": "224.685",
"value": "Boldklubben 1908_sentence_1"
},
{
"operation": "Highlighting",
"time": "232.037",
"value": "Boldklubben 1908_sentence_4"
},
{
"operation": "Highlighting",
"time": "233.507",
"value": "Boldklubben 1908_sentence_5"
},
{
"operation": "Highlighting",
"time": "234.135",
"value": "Boldklubben 1908_sentence_6"
},
{
"operation": "search",
"time": "276.091",
"value": "2011 boldklubben 1908"
},
{
"operation": "Now on",
"time": "278.353",
"value": "undefined"
},
{
"operation": "search",
"time": "287.167",
"value": "List of Boldklubben 1908 seasons"
},
{
"operation": "Now on",
"time": "289.195",
"value": "List of Boldklubben 1908 seasons"
},
{
"operation": "Highlighting",
"time": "291.3",
"value": "List of Boldklubben 1908 seasons_sentence_0"
},
{
"operation": "Highlighting",
"time": "295.068",
"value": "List of Boldklubben 1908 seasons_sentence_5"
},
{
"operation": "hyperlink",
"time": "314.346",
"value": "B 1908"
},
{
"operation": "Now on",
"time": "315.125",
"value": "Boldklubben 1908"
},
{
"operation": "finish",
"time": "335.497",
"value": "finish"
}
]
}
|
74149
|
John William Inchbold, who was born on April 29, 1830 at Yorkshire, has shown early talent in drawing but refused to become a draughtsman in the lithographic works of Day and Hughe.
|
{
"evidence": [
"John William Inchbold_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:\nJohn William Inchbold, who was born on April 29, 1830 at Yorkshire, has shown early talent in drawing but refused to become a draughtsman in the lithographic works of Day and Hughe.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"John William Inchbold_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6866,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.343",
"value": "?search="
},
{
"operation": "search",
"time": "40.609",
"value": "John William Inchbold"
},
{
"operation": "Now on",
"time": "42.104",
"value": "John William Inchbold"
},
{
"operation": "Highlighting",
"time": "48.027",
"value": "John William Inchbold_sentence_0"
},
{
"operation": "finish",
"time": "62.142",
"value": "finish"
}
]
}
|
43841
|
Find My (or similar is the name given by various manufacturers to software and a service for smartphones), which Apple offers as a free service, helps to locate lost or stolen phones.
|
{
"evidence": [
"Find My Phone_sentence_0",
"Find My Phone_sentence_2",
"Find My Phone_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:\nFind My (or similar is the name given by various manufacturers to software and a service for smartphones), which Apple offers as a free service, helps to locate lost or stolen phones.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Find My Phone_sentence_0",
"Find My Phone_sentence_2",
"Find My Phone_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6867,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.73",
"value": "?search="
},
{
"operation": "search",
"time": "5.857",
"value": "Find My"
},
{
"operation": "Now on",
"time": "6.431",
"value": "Find My"
},
{
"operation": "Page search",
"time": "12.154",
"value": "smartphones"
},
{
"operation": "Page search",
"time": "12.353",
"value": "smartphones"
},
{
"operation": "Page search",
"time": "12.532",
"value": "smartphones"
},
{
"operation": "Page search",
"time": "12.731",
"value": "smartphones"
},
{
"operation": "Page search",
"time": "13.158",
"value": "smartphones"
},
{
"operation": "Page search",
"time": "13.409",
"value": "smartphones"
},
{
"operation": "Page search",
"time": "22.742",
"value": "free service"
},
{
"operation": "Page search",
"time": "22.934",
"value": "free service"
},
{
"operation": "Page search",
"time": "23.08",
"value": "free service"
},
{
"operation": "Page search",
"time": "23.247",
"value": "free service"
},
{
"operation": "Page search",
"time": "23.446",
"value": "free service"
},
{
"operation": "Page search",
"time": "23.629",
"value": "free service"
},
{
"operation": "Page search",
"time": "23.798",
"value": "free servic"
},
{
"operation": "Page search",
"time": "23.973",
"value": "free servi"
},
{
"operation": "Page search",
"time": "24.132",
"value": "free serv"
},
{
"operation": "Page search",
"time": "24.305",
"value": "free ser"
},
{
"operation": "Page search",
"time": "24.467",
"value": "free se"
},
{
"operation": "Page search",
"time": "24.631",
"value": "free s"
},
{
"operation": "Page search",
"time": "24.807",
"value": "free s"
},
{
"operation": "Page search",
"time": "24.974",
"value": "free"
},
{
"operation": "Page search",
"time": "25.164",
"value": "free"
},
{
"operation": "Page search",
"time": "25.34",
"value": "free"
},
{
"operation": "Page search",
"time": "25.523",
"value": "fre"
},
{
"operation": "Page search",
"time": "25.72",
"value": "fr"
},
{
"operation": "Page search",
"time": "25.892",
"value": "fr"
},
{
"operation": "page-search-reset",
"time": "26.09",
"value": "page-search-reset"
},
{
"operation": "page-search-reset",
"time": "26.273",
"value": "page-search-reset"
},
{
"operation": "page-search-reset",
"time": "26.541",
"value": "page-search-reset"
},
{
"operation": "search",
"time": "39.111",
"value": "Find My Phone"
},
{
"operation": "Now on",
"time": "39.808",
"value": "Find My Phone"
},
{
"operation": "Page search",
"time": "51.278",
"value": "name given"
},
{
"operation": "Page search",
"time": "51.488",
"value": "name given"
},
{
"operation": "Highlighting",
"time": "52.673",
"value": "Find My Phone_sentence_0"
},
{
"operation": "Page search",
"time": "58.423",
"value": "free service"
},
{
"operation": "Page search",
"time": "58.603",
"value": "free service"
},
{
"operation": "Page search",
"time": "58.765",
"value": "free service"
},
{
"operation": "Highlighting",
"time": "60.156",
"value": "Find My Phone_sentence_2"
},
{
"operation": "Highlighting",
"time": "70.353",
"value": "Find My Phone_sentence_1"
},
{
"operation": "finish",
"time": "306.995",
"value": "finish"
}
]
}
|
29825
|
In 1963, Gam was a leading member of The Minnesota Theatre Company in the opening season of The Tyrone Guthrie Theater in Minneapolis; then in 2003, she refused to appear in the rotating cast of the Off-Broadway stage reading of Wit & Wisdom.
|
{
"evidence": [
"Rita Gam_sentence_15"
],
"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:\nIn 1963, Gam was a leading member of The Minnesota Theatre Company in the opening season of The Tyrone Guthrie Theater in Minneapolis; then in 2003, she refused to appear in the rotating cast of the Off-Broadway stage reading of Wit & Wisdom.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Rita Gam_sentence_15"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6868,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.625",
"value": "?search="
},
{
"operation": "search",
"time": "51.805",
"value": "Rita Gam"
},
{
"operation": "Now on",
"time": "53.017",
"value": "Rita Gam"
},
{
"operation": "Highlighting",
"time": "61.707",
"value": "Rita Gam_sentence_12"
},
{
"operation": "Highlighting",
"time": "72.507",
"value": "Rita Gam_sentence_15"
},
{
"operation": "Highlighting",
"time": "77.728",
"value": "Rita Gam_sentence_10"
},
{
"operation": "Highlighting deleted",
"time": "79.507",
"value": "Rita Gam_sentence_10"
},
{
"operation": "finish",
"time": "86.284",
"value": "finish"
}
]
}
|
exfever_3240
|
Cleopatra is a 1964 American epic historical drama film directed by Joseph L. Mankiewicz, with a screenplay adapted by an American screenwriter who scripted such films as Mildred Pierce (1945) and The Unsuspected (1947).
|
{
"evidence": [
"Cleopatra (1963 film)",
"Ranald MacDougall"
],
"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:\nCleopatra is a 1964 American epic historical drama film directed by Joseph L. Mankiewicz, with a screenplay adapted by an American screenwriter who scripted such films as Mildred Pierce (1945) and The Unsuspected (1947).\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Cleopatra (1963 film)",
"Ranald MacDougall"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6869,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
38040
|
Cyrus Smith, one of the protagonists of Jules Verne's 1875 novel The Mysterious Island, is a selfless and courageous man with practical knowledge of physics, chemistry, and botany.
|
{
"evidence": [
"Cyrus Smith_sentence_3",
"Cyrus Smith_sentence_7",
"Cyrus Smith_sentence_8"
],
"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:\nCyrus Smith, one of the protagonists of Jules Verne's 1875 novel The Mysterious Island, is a selfless and courageous man with practical knowledge of physics, chemistry, and botany.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Cyrus Smith_sentence_3",
"Cyrus Smith_sentence_7",
"Cyrus Smith_sentence_8"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6870,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.757",
"value": "?search="
},
{
"operation": "search",
"time": "15.006",
"value": "Cyrus Smith"
},
{
"operation": "Now on",
"time": "16.017",
"value": "Cyrus Smith"
},
{
"operation": "Highlighting",
"time": "21.598",
"value": "Cyrus Smith_sentence_3"
},
{
"operation": "Highlighting",
"time": "31.512",
"value": "Cyrus Smith_sentence_7"
},
{
"operation": "Highlighting",
"time": "33.084",
"value": "Cyrus Smith_sentence_8"
},
{
"operation": "finish",
"time": "39.27",
"value": "finish"
}
]
}
|
exfever_544
|
Elise Demetria Neal is an American actress who starred as Yvonne Hughley in the Hughleys sitcom on an Italian broadcast television network that launched on January 16, 1995.
|
{
"evidence": [
"Elise Neal",
"UPN"
],
"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:\nElise Demetria Neal is an American actress who starred as Yvonne Hughley in the Hughleys sitcom on an Italian broadcast television network that launched on January 16, 1995.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Elise Neal",
"UPN"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6871,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
72832
|
In late 2008 and early 2009, the Snuggie brand of sleeved blankets became a pop culture phenomenon, then it was featured on television programs like Today where cast and crew donned Snuggie blankets for a segment which was described as looking like a Black Protestant choir.
|
{
"evidence": [
"Sleeved blanket_sentence_5",
"Sleeved blanket_sentence_7"
],
"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:\nIn late 2008 and early 2009, the Snuggie brand of sleeved blankets became a pop culture phenomenon, then it was featured on television programs like Today where cast and crew donned Snuggie blankets for a segment which was described as looking like a Black Protestant choir.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Sleeved blanket_sentence_5",
"Sleeved blanket_sentence_7"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6872,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.913",
"value": "?search="
},
{
"operation": "search",
"time": "25.976",
"value": "Snuggie"
},
{
"operation": "Now on",
"time": "26.732",
"value": "Sleeved blanket"
},
{
"operation": "Highlighting",
"time": "181.767",
"value": "Sleeved blanket_sentence_5"
},
{
"operation": "Highlighting",
"time": "189.214",
"value": "Sleeved blanket_sentence_7"
},
{
"operation": "finish",
"time": "204.869",
"value": "finish"
}
]
}
|
exfever_2531
|
MacArthur became the most decorated American soldier of the war for his contribution to winning the war.
|
{
"evidence": [
"Thomas Pynchon",
"Gravity's Rainbow"
],
"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:\nMacArthur became the most decorated American soldier of the war for his contribution to winning the war.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Thomas Pynchon",
"Gravity's Rainbow"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6873,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
25801
|
Joanne Tseng has an older sister, her parents are teachers and her family is part of the Atayal which are about 90,000 people.
|
{
"evidence": [
"Joanne Tseng_sentence_3",
"Joanne Tseng_sentence_4",
"Joanne Tseng_sentence_5",
"Atayal people_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:\nJoanne Tseng has an older sister, her parents are teachers and her family is part of the Atayal which are about 90,000 people.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Joanne Tseng_sentence_3",
"Joanne Tseng_sentence_4",
"Joanne Tseng_sentence_5",
"Atayal people_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6874,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.142",
"value": "?search="
},
{
"operation": "search",
"time": "13.701",
"value": "Joanne Tseng"
},
{
"operation": "Now on",
"time": "15.88",
"value": "Joanne Tseng"
},
{
"operation": "Highlighting",
"time": "28.29",
"value": "Joanne Tseng_sentence_3"
},
{
"operation": "Highlighting",
"time": "30.594",
"value": "Joanne Tseng_sentence_4"
},
{
"operation": "Highlighting",
"time": "34.59",
"value": "Joanne Tseng_sentence_5"
},
{
"operation": "hyperlink",
"time": "39.011",
"value": "Atayal"
},
{
"operation": "Now on",
"time": "40.307",
"value": "Atayal people"
},
{
"operation": "Highlighting",
"time": "45.341",
"value": "Atayal people_sentence_3"
},
{
"operation": "finish",
"time": "64.005",
"value": "finish"
}
]
}
|
37938
|
Denisa Šátralová is a Czech tennis player who was born in Prague, the capital and the largest city in Czech Republic.
|
{
"evidence": [
"Denisa Šátralová_sentence_0",
"Denisa Šátralová_sentence_4",
"Prague_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:\nDenisa Šátralová is a Czech tennis player who was born in Prague, the capital and the largest city in Czech Republic.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Denisa Šátralová_sentence_0",
"Denisa Šátralová_sentence_4",
"Prague_sentence_4"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6875,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.16",
"value": "?search="
},
{
"operation": "search",
"time": "12.421",
"value": "Denisa Šátralová"
},
{
"operation": "Now on",
"time": "13.359",
"value": "Denisa Šátralová"
},
{
"operation": "Highlighting",
"time": "16.292",
"value": "Denisa Šátralová_sentence_0"
},
{
"operation": "Highlighting",
"time": "22.79",
"value": "Denisa Šátralová_sentence_4"
},
{
"operation": "hyperlink",
"time": "25.84",
"value": "Prague"
},
{
"operation": "Now on",
"time": "27.236",
"value": "Prague"
},
{
"operation": "Highlighting",
"time": "31.558",
"value": "Prague_sentence_4"
},
{
"operation": "finish",
"time": "39.442",
"value": "finish"
}
]
}
|
exfever_1503
|
Tobias Vincent Maguire is an American actor and film producer best known for playing the title character in the Spider-Man trilogy by an American filmmaker, producer, and actor known for directing a few movies, including a neo-noir crime-thriller television series set in rural Minnesota that earned multiple awards and nominations.
|
{
"evidence": [
"Tobey Maguire",
"Sam Raimi",
"A Simple Plan (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:\nTobias Vincent Maguire is an American actor and film producer best known for playing the title character in the Spider-Man trilogy by an American filmmaker, producer, and actor known for directing a few movies, including a neo-noir crime-thriller television series set in rural Minnesota that earned multiple awards and nominations.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tobey Maguire",
"Sam Raimi",
"A Simple Plan (film)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6876,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_357
|
Geraldine Leigh Chaplin is an actress who came to prominence in her portrayal of Tonya in a 1965 epic historical romance film directed by David Lean.
|
{
"evidence": [
"Geraldine Chaplin",
"Doctor Zhivago (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:\nGeraldine Leigh Chaplin is an actress who came to prominence in her portrayal of Tonya in a 1965 epic historical romance film directed by David Lean.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Geraldine Chaplin",
"Doctor Zhivago (film)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6877,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
45085
|
In 1902, the Clemson Tigers football team, representing the Clemson Tigers of Clemson Agricultural College, didn't play North Carolina A&M, but played Georgia Tech on October 18.
|
{
"evidence": [
"1902 Clemson Tigers football team_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:\nIn 1902, the Clemson Tigers football team, representing the Clemson Tigers of Clemson Agricultural College, didn't play North Carolina A&M, but played Georgia Tech on October 18.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"1902 Clemson Tigers football team_sentence_7"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6878,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.762",
"value": "?search="
},
{
"operation": "search",
"time": "16.211",
"value": "1902 Clemson Tigers football team"
},
{
"operation": "Now on",
"time": "17.508",
"value": "1902 Clemson Tigers football team"
},
{
"operation": "Highlighting",
"time": "20.287",
"value": "1902 Clemson Tigers football team_sentence_0"
},
{
"operation": "Highlighting",
"time": "33.361",
"value": "1902 Clemson Tigers football team_sentence_7"
},
{
"operation": "finish",
"time": "62.926",
"value": "finish"
}
]
}
|
exfever_8308
|
Backbeat is Sheryl Lee's seventh film.
|
{
"evidence": [
"Sheryl Lee",
"Twin Peaks",
"American Broadcasting Company"
],
"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:\nBackbeat is Sheryl Lee's seventh film.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Sheryl Lee",
"Twin Peaks",
"American Broadcasting Company"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6879,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
66291
|
Aminé released his songs entitled "Warm on a Cold Night" (Remix), "Redmercedes (Remix)", "Blinds", and "Blackjack (Remix)" in 2017.
|
{
"evidence": [
"Good for You (album)_sentence_0",
"Good for You (album)_sentence_3",
"Good for You (album)_sentence_1",
"Aminé (rapper)_sentence_19"
],
"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:\nAminé released his songs entitled \"Warm on a Cold Night\" (Remix), \"Redmercedes (Remix)\", \"Blinds\", and \"Blackjack (Remix)\" in 2017.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Good for You (album)_sentence_0",
"Good for You (album)_sentence_3",
"Good for You (album)_sentence_1",
"Aminé (rapper)_sentence_19"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6880,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.878",
"value": "?search="
},
{
"operation": "search",
"time": "18.83",
"value": "Aminé"
},
{
"operation": "Now on",
"time": "26.448",
"value": "Aminé"
},
{
"operation": "hyperlink",
"time": "82.174",
"value": "Good for You (album)"
},
{
"operation": "Now on",
"time": "82.926",
"value": "Good for You (album)"
},
{
"operation": "Highlighting",
"time": "86.55",
"value": "Good for You (album)_sentence_0"
},
{
"operation": "hyperlink",
"time": "130.294",
"value": "2 Singles Good for You (album)_section_1"
},
{
"operation": "Highlighting",
"time": "150.862",
"value": "Good for You (album)_sentence_3"
},
{
"operation": "Highlighting",
"time": "164.821",
"value": "Good for You (album)_sentence_1"
},
{
"operation": "back-button-clicked",
"time": "172.478",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "173.116",
"value": "Aminé (rapper)"
},
{
"operation": "Highlighting",
"time": "185.165",
"value": "Aminé (rapper)_sentence_19"
},
{
"operation": "finish",
"time": "239.741",
"value": "finish"
}
]
}
|
25570
|
Gopala Dasa, named "Vijaya Dasa" at birth, was a prominent 18th-century Kannada language poet and saint belonging to the Haridasa tradition.
|
{
"evidence": [
"Gopala Dasa_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:\nGopala Dasa, named \"Vijaya Dasa\" at birth, was a prominent 18th-century Kannada language poet and saint belonging to the Haridasa tradition.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Gopala Dasa_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6881,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.589",
"value": "?search="
},
{
"operation": "search",
"time": "10.923",
"value": "Contains...Gopala Dasa"
},
{
"operation": "Now on",
"time": "11.852",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "14.554",
"value": "Gopala Dasa"
},
{
"operation": "Now on",
"time": "15.07",
"value": "Gopala Dasa"
},
{
"operation": "Page search",
"time": "56.921",
"value": "Vijaya"
},
{
"operation": "Page search",
"time": "91.646",
"value": "birth"
},
{
"operation": "Highlighting",
"time": "95.02",
"value": "Gopala Dasa_sentence_2"
},
{
"operation": "finish",
"time": "115.636",
"value": "finish"
}
]
}
|
exfever_7073
|
Dame Julia Mary Walters is an English actress, author, and comedian, who rose to prominence by playing the title role in a 1983 British film whose genre of dramatic work combines elements of comedy and action.
|
{
"evidence": [
"Julie Walters",
"Educating Rita (film)",
"Comedy-drama"
],
"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:\nDame Julia Mary Walters is an English actress, author, and comedian, who rose to prominence by playing the title role in a 1983 British film whose genre of dramatic work combines elements of comedy and action.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Julie Walters",
"Educating Rita (film)",
"Comedy-drama"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6882,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
12719
|
Henry Martyn Scudder founded the American Arcot Mission in 1850 before his health failed in 1864.
|
{
"evidence": [
"Henry Martyn Scudder_sentence_10",
"Henry Martyn Scudder_sentence_12"
],
"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:\nHenry Martyn Scudder founded the American Arcot Mission in 1850 before his health failed in 1864.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Henry Martyn Scudder_sentence_10",
"Henry Martyn Scudder_sentence_12"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6883,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.471",
"value": "?search="
},
{
"operation": "search",
"time": "43.975",
"value": "Henry Martyn Scudder"
},
{
"operation": "Now on",
"time": "45.3",
"value": "Henry Martyn Scudder"
},
{
"operation": "Highlighting",
"time": "53.693",
"value": "Henry Martyn Scudder_sentence_1"
},
{
"operation": "Highlighting",
"time": "70.316",
"value": "Henry Martyn Scudder_sentence_10"
},
{
"operation": "Page search",
"time": "107.178",
"value": "health"
},
{
"operation": "Highlighting",
"time": "135.416",
"value": "Henry Martyn Scudder_sentence_12"
},
{
"operation": "finish",
"time": "175.496",
"value": "finish"
}
]
}
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.