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_5234
|
Stone Gossard is a record producer.
|
{
"evidence": [
"Pearl Jam",
"Eddie Vedder",
"Rolling Stone"
],
"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:\nStone Gossard is a record producer.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Pearl Jam",
"Eddie Vedder",
"Rolling Stone"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6984,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7705
|
Logan is a 2017 superhero film and the third and final installment in the Wolverine trilogy following a 2009 American superhero film written by an American writer, director, and producer.
|
{
"evidence": [
"Logan (film)",
"X-Men Origins: Wolverine",
"David Benioff"
],
"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:\nLogan is a 2017 superhero film and the third and final installment in the Wolverine trilogy following a 2009 American superhero film written by an American writer, director, and producer.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Logan (film)",
"X-Men Origins: Wolverine",
"David Benioff"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6985,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5597
|
Doctor Stephen Strange is a fictional character appearing in a thick book, averaging 32 pages and containing comics published by Marvel Comics.
|
{
"evidence": [
"Doctor Strange",
"American comic book"
],
"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:\nDoctor Stephen Strange is a fictional character appearing in a thick book, averaging 32 pages and containing comics published by Marvel Comics.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Doctor Strange",
"American comic book"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6986,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
56452
|
Jean Marie Auguste Bachelot de La Pylaie (May 25, 1786 in France- October 12, 1856) was a botanist, explorer, archaeologist, author of various archaeological studies and first known collector of local species in Saint Pierre who studied at Laval and then in Paris at the Muséum national d'histoire naturelle.
|
{
"evidence": [
"Jean Bachelot La Pylaie_sentence_0",
"Jean Bachelot La Pylaie_sentence_7",
"Jean Bachelot La Pylaie_sentence_4",
"Jean Bachelot La Pylaie_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:\nJean Marie Auguste Bachelot de La Pylaie (May 25, 1786 in France- October 12, 1856) was a botanist, explorer, archaeologist, author of various archaeological studies and first known collector of local species in Saint Pierre who studied at Laval and then in Paris at the Muséum national d'histoire naturelle.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Jean Bachelot La Pylaie_sentence_0",
"Jean Bachelot La Pylaie_sentence_7",
"Jean Bachelot La Pylaie_sentence_4",
"Jean Bachelot La Pylaie_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6987,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.913",
"value": "?search="
},
{
"operation": "search",
"time": "10.161",
"value": "Jean Marie Auguste Bachelot de La Pylai"
},
{
"operation": "Now on",
"time": "11.054",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "14.284",
"value": "jean marie auguste bachelor de la pylaie"
},
{
"operation": "Now on",
"time": "14.906",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "19.935",
"value": "jean marie auguste bachelor de la pylaie"
},
{
"operation": "Now on",
"time": "20.641",
"value": "undefined"
},
{
"operation": "search",
"time": "31.023",
"value": "ean Marie Auguste Bachelot"
},
{
"operation": "Now on",
"time": "31.629",
"value": "undefined"
},
{
"operation": "search",
"time": "130.094",
"value": "Jean Bachelot La Pylaie"
},
{
"operation": "Now on",
"time": "130.937",
"value": "Jean Bachelot La Pylaie"
},
{
"operation": "Highlighting",
"time": "141.801",
"value": "Jean Bachelot La Pylaie_sentence_0"
},
{
"operation": "Highlighting",
"time": "157.593",
"value": "Jean Bachelot La Pylaie_sentence_7"
},
{
"operation": "Highlighting",
"time": "168.009",
"value": "Jean Bachelot La Pylaie_sentence_4"
},
{
"operation": "Highlighting",
"time": "176.715",
"value": "Jean Bachelot La Pylaie_sentence_1"
},
{
"operation": "finish",
"time": "199.93",
"value": "finish"
}
]
}
|
21042
|
Babylon Mystery Religion is now out of print and a second book is available entitled "The Babylon Connection?" in which Woodrow reiterates his views previously presented in Babylon Mystery Religion.
|
{
"evidence": [
"Babylon Mystery Religion_sentence_3",
"Babylon Mystery Religion_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:\nBabylon Mystery Religion is now out of print and a second book is available entitled \"The Babylon Connection?\" in which Woodrow reiterates his views previously presented in Babylon Mystery Religion.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Babylon Mystery Religion_sentence_3",
"Babylon Mystery Religion_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6988,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.669",
"value": "?search="
},
{
"operation": "search",
"time": "88.282",
"value": "Babylon Mystery Religion"
},
{
"operation": "Now on",
"time": "88.959",
"value": "Babylon Mystery Religion"
},
{
"operation": "Highlighting",
"time": "107.649",
"value": "Babylon Mystery Religion_sentence_3"
},
{
"operation": "Highlighting",
"time": "134.216",
"value": "Babylon Mystery Religion_sentence_4"
},
{
"operation": "finish",
"time": "142.805",
"value": "finish"
}
]
}
|
16846
|
The East Africa Premier League, which was founded in 2011, is participated by four teams from Kenya and two teams from Uganda.
|
{
"evidence": [
"East Africa Premier League_sentence_1",
"East Africa Premier League_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe East Africa Premier League, which was founded in 2011, is participated by four teams from Kenya and two teams from Uganda.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"East Africa Premier League_sentence_1",
"East Africa Premier League_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6989,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.661",
"value": "?search="
},
{
"operation": "search",
"time": "12.471",
"value": "East Africa Premier League"
},
{
"operation": "Now on",
"time": "13.714",
"value": "East Africa Premier League"
},
{
"operation": "Highlighting",
"time": "24.126",
"value": "East Africa Premier League_sentence_1"
},
{
"operation": "Highlighting",
"time": "32.297",
"value": "East Africa Premier League_sentence_2"
},
{
"operation": "finish",
"time": "63.184",
"value": "finish"
}
]
}
|
exfever_1378
|
How I Met Your Mother is an American sitcom about a group of friends in the most deserted city in the United States.
|
{
"evidence": [
"How I Met Your Mother",
"New York City"
],
"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:\nHow I Met Your Mother is an American sitcom about a group of friends in the most deserted city in the United States.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"How I Met Your Mother",
"New York City"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6990,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5357
|
Starrcade was regarded by the NWA and WCW as their flagship event of the year, much in the same vein that its rival, the World Wrestling Federation, regarded the professional wrestling live event held annually between mid-March and mid-April.
|
{
"evidence": [
"Starrcade",
"WrestleMania"
],
"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:\nStarrcade was regarded by the NWA and WCW as their flagship event of the year, much in the same vein that its rival, the World Wrestling Federation, regarded the professional wrestling live event held annually between mid-March and mid-April.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Starrcade",
"WrestleMania"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6991,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7986
|
Alicia Christian "Jodie" Foster has received various awards, including three BAFTA Film Awards, which were initially held in the pedestrianized area in the West End of London, England, at the flagship Odeon theater.
|
{
"evidence": [
"Jodie Foster",
"British Academy Film Awards",
"Leicester Square"
],
"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:\nAlicia Christian \"Jodie\" Foster has received various awards, including three BAFTA Film Awards, which were initially held in the pedestrianized area in the West End of London, England, at the flagship Odeon theater.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Jodie Foster",
"British Academy Film Awards",
"Leicester Square"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6992,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
14989
|
Cullingworth is a village and civil parish 7 miles west of Bradford and 3 miles south of Keighley, 2 miles to the east of Haworth and is situated near the head of Harden valley in West Yorkshire, England.
|
{
"evidence": [
"Cullingworth_sentence_0",
"Cullingworth_sentence_1",
"Haworth_sentence_1",
"Cullingworth_sentence_4",
"Keighley_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:\nCullingworth is a village and civil parish 7 miles west of Bradford and 3 miles south of Keighley, 2 miles to the east of Haworth and is situated near the head of Harden valley in West Yorkshire, England.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Cullingworth_sentence_0",
"Cullingworth_sentence_1",
"Haworth_sentence_1",
"Cullingworth_sentence_4",
"Keighley_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6993,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.724",
"value": "?search="
},
{
"operation": "search",
"time": "14.692",
"value": "Cullingworth"
},
{
"operation": "Now on",
"time": "16.208",
"value": "Cullingworth"
},
{
"operation": "Highlighting",
"time": "24.966",
"value": "Cullingworth_sentence_0"
},
{
"operation": "Highlighting",
"time": "31.872",
"value": "Cullingworth_sentence_1"
},
{
"operation": "Page search",
"time": "86.315",
"value": "haworth"
},
{
"operation": "search",
"time": "108.693",
"value": "Haworth"
},
{
"operation": "Now on",
"time": "110.702",
"value": "Haworth"
},
{
"operation": "Highlighting",
"time": "129.747",
"value": "Haworth_sentence_1"
},
{
"operation": "search",
"time": "166.204",
"value": "west yorkshire"
},
{
"operation": "Now on",
"time": "168.66",
"value": "West Yorkshire"
},
{
"operation": "search",
"time": "191.749",
"value": "harden valley"
},
{
"operation": "Now on",
"time": "193.25",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "212.174",
"value": "hardin valley"
},
{
"operation": "Now on",
"time": "213.843",
"value": "undefined"
},
{
"operation": "search",
"time": "222.04",
"value": "Cullingworth"
},
{
"operation": "Now on",
"time": "223.241",
"value": "Cullingworth"
},
{
"operation": "Highlighting",
"time": "239.754",
"value": "Cullingworth_sentence_4"
},
{
"operation": "hyperlink",
"time": "253.286",
"value": "Keighley"
},
{
"operation": "Now on",
"time": "254.31",
"value": "Keighley railway station"
},
{
"operation": "back-button-clicked",
"time": "271.648",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "271.793",
"value": "Cullingworth"
},
{
"operation": "hyperlink",
"time": "276.783",
"value": "Keighley"
},
{
"operation": "Now on",
"time": "279.946",
"value": "Keighley"
},
{
"operation": "Highlighting",
"time": "285.963",
"value": "Keighley_sentence_1"
},
{
"operation": "finish",
"time": "306.265",
"value": "finish"
}
]
}
|
83904
|
Wild Wild Country, which won the Creative Arts Emmy Award for Outstanding Documentary or Nonfiction Series in 2018, received an average rating of 9.92 on Rotten Tomatoes.
|
{
"evidence": [
"Wild Wild Country_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:\nWild Wild Country, which won the Creative Arts Emmy Award for Outstanding Documentary or Nonfiction Series in 2018, received an average rating of 9.92 on Rotten Tomatoes.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Wild Wild Country_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6994,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.125",
"value": "?search="
},
{
"operation": "search",
"time": "12.865",
"value": "Wild Wild Country"
},
{
"operation": "Now on",
"time": "13.684",
"value": "Wild Wild Country"
},
{
"operation": "Highlighting",
"time": "34.397",
"value": "Wild Wild Country_sentence_5"
},
{
"operation": "Highlighting",
"time": "54.254",
"value": "Wild Wild Country_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "58.041",
"value": "Wild Wild Country_sentence_5"
},
{
"operation": "finish",
"time": "69.766",
"value": "finish"
}
]
}
|
33667
|
The Grand Morbid Funeral was recorded in the 90s in Ireland and was released in November 1999 by Universal Records.
|
{
"evidence": [
"Grand Morbid Funeral_sentence_2"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Grand Morbid Funeral was recorded in the 90s in Ireland and was released in November 1999 by Universal Records.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Grand Morbid Funeral_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6995,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.603",
"value": "?search="
},
{
"operation": "search",
"time": "16.061",
"value": "Contains...The Grand Morbid Funeral"
},
{
"operation": "Now on",
"time": "17.107",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "19.61",
"value": "Grand Morbid Funeral"
},
{
"operation": "Now on",
"time": "20.11",
"value": "Grand Morbid Funeral"
},
{
"operation": "Highlighting",
"time": "32.56",
"value": "Grand Morbid Funeral_sentence_2"
},
{
"operation": "finish",
"time": "40.96",
"value": "finish"
}
]
}
|
62009
|
During the 2015 Sam Houston State Bearkats football team match against the Stephen F. Austin State Lumberjacks (the 90th time the teams faced each other), the two teams had a back and forth during the first 2 quarters ending in a score of 10-3 before half time, with SFA leading.
|
{
"evidence": [
"2015 Sam Houston State Bearkats football team_sentence_18"
],
"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:\nDuring the 2015 Sam Houston State Bearkats football team match against the Stephen F. Austin State Lumberjacks (the 90th time the teams faced each other), the two teams had a back and forth during the first 2 quarters ending in a score of 10-3 before half time, with SFA leading.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"2015 Sam Houston State Bearkats football team_sentence_18"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6996,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.089",
"value": "?search="
},
{
"operation": "search",
"time": "56.163",
"value": "2015 Sam Houston State Bearkats football team"
},
{
"operation": "Now on",
"time": "57.2",
"value": "2015 Sam Houston State Bearkats football team"
},
{
"operation": "Highlighting",
"time": "101.033",
"value": "2015 Sam Houston State Bearkats football team_cell_1_4_2"
},
{
"operation": "Highlighting",
"time": "101.034",
"value": "2015 Sam Houston State Bearkats football team_section_0"
},
{
"operation": "Highlighting",
"time": "101.037",
"value": "2015 Sam Houston State Bearkats football team_header_cell_1_0_2"
},
{
"operation": "Highlighting deleted",
"time": "104.674",
"value": "2015 Sam Houston State Bearkats football team_cell_1_4_2"
},
{
"operation": "Highlighting deleted",
"time": "104.678",
"value": "2015 Sam Houston State Bearkats football team_section_0"
},
{
"operation": "Highlighting deleted",
"time": "104.682",
"value": "2015 Sam Houston State Bearkats football team_header_cell_1_0_2"
},
{
"operation": "hyperlink",
"time": "133.219",
"value": "2.4 Stephen F. Austin 2015 Sam Houston State Bearkats football team_section_5"
},
{
"operation": "Highlighting",
"time": "136.409",
"value": "2015 Sam Houston State Bearkats football team_sentence_12"
},
{
"operation": "Highlighting",
"time": "171.425",
"value": "2015 Sam Houston State Bearkats football team_sentence_14"
},
{
"operation": "Highlighting",
"time": "175.586",
"value": "2015 Sam Houston State Bearkats football team_sentence_15"
},
{
"operation": "Highlighting",
"time": "180.171",
"value": "2015 Sam Houston State Bearkats football team_sentence_16"
},
{
"operation": "Highlighting",
"time": "235.347",
"value": "2015 Sam Houston State Bearkats football team_sentence_18"
},
{
"operation": "finish",
"time": "263.108",
"value": "finish"
}
]
}
|
22914
|
Nif gene are genes encoding enzymes involved in the fixation of atmospheric nitrogen into a form of nitrogen available to living organisms and are found in both free-living nitrogen-fixing bacteria and in symbiotic bacteria, also referred as diazotroph, associated with various plants.
|
{
"evidence": [
"Nif gene_sentence_0",
"Nif gene_sentence_3",
"Diazotroph_sentence_0"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nNif gene are genes encoding enzymes involved in the fixation of atmospheric nitrogen into a form of nitrogen available to living organisms and are found in both free-living nitrogen-fixing bacteria and in symbiotic bacteria, also referred as diazotroph, associated with various plants.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Nif gene_sentence_0",
"Nif gene_sentence_3",
"Diazotroph_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6997,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.857",
"value": "?search="
},
{
"operation": "search",
"time": "15.644",
"value": "Nif gene"
},
{
"operation": "Now on",
"time": "16.728",
"value": "Nif gene"
},
{
"operation": "Highlighting",
"time": "22.139",
"value": "Nif gene_sentence_0"
},
{
"operation": "Highlighting",
"time": "36.416",
"value": "Nif gene_sentence_3"
},
{
"operation": "hyperlink",
"time": "51.137",
"value": "free-living nitrogen-fixing bacteria and in symbiotic bacteria"
},
{
"operation": "Now on",
"time": "51.854",
"value": "Diazotrophs"
},
{
"operation": "Highlighting",
"time": "60.081",
"value": "Diazotroph_sentence_0"
},
{
"operation": "finish",
"time": "165.566",
"value": "finish"
}
]
}
|
32859
|
Matchday two happened on July 28, 2009 against Eintracht Braunschweig, then Matchday three happened on August 7, 2009 against Werder Bremen II.
|
{
"evidence": [
"2009–10 Borussia Dortmund II season_sentence_5",
"2009–10 Borussia Dortmund II season_sentence_7",
"2009–10 Borussia Dortmund II season_sentence_5",
"2009–10 SV Werder Bremen II season_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:\nMatchday two happened on July 28, 2009 against Eintracht Braunschweig, then Matchday three happened on August 7, 2009 against Werder Bremen II.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"2009–10 Borussia Dortmund II season_sentence_5",
"2009–10 Borussia Dortmund II season_sentence_7",
"2009–10 Borussia Dortmund II season_sentence_5",
"2009–10 SV Werder Bremen II season_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 6998,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.679",
"value": "?search="
},
{
"operation": "search",
"time": "28.311",
"value": "Eintracht Braunschweig"
},
{
"operation": "Now on",
"time": "29.546",
"value": "Eintracht Braunschweig"
},
{
"operation": "search",
"time": "41.501",
"value": "2009 Eintracht Braunschweig"
},
{
"operation": "Now on",
"time": "42.658",
"value": "undefined"
},
{
"operation": "search",
"time": "65.91",
"value": "Werder Bremen II 2009"
},
{
"operation": "Now on",
"time": "67.428",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "69.381",
"value": "2009–10 SV Werder Bremen II season"
},
{
"operation": "Now on",
"time": "70.486",
"value": "2009–10 SV Werder Bremen II season"
},
{
"operation": "hyperlink",
"time": "76.372",
"value": "1.1.1 July and August 2009–10 SV Werder Bremen II season_section_2"
},
{
"operation": "Highlighting",
"time": "83.014",
"value": "2009–10 SV Werder Bremen II season_sentence_5"
},
{
"operation": "Highlighting deleted",
"time": "85.309",
"value": "2009–10 SV Werder Bremen II season_sentence_5"
},
{
"operation": "hyperlink",
"time": "86.391",
"value": "Borussia Dortmund II"
},
{
"operation": "Now on",
"time": "87.165",
"value": "Borussia Dortmund II"
},
{
"operation": "search",
"time": "97.751",
"value": "Contains...2009 Borussia Dortmund II"
},
{
"operation": "Now on",
"time": "98.597",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "99.932",
"value": "2009–10 Borussia Dortmund II season"
},
{
"operation": "Now on",
"time": "101.025",
"value": "2009–10 Borussia Dortmund II season"
},
{
"operation": "hyperlink",
"time": "102.108",
"value": "1.1.1 July and August 2009–10 Borussia Dortmund II season_section_2"
},
{
"operation": "Highlighting",
"time": "105.909",
"value": "2009–10 Borussia Dortmund II season_sentence_5"
},
{
"operation": "Highlighting",
"time": "110.1",
"value": "2009–10 Borussia Dortmund II season_sentence_7"
},
{
"operation": "hyperlink",
"time": "128.685",
"value": "Eintracht Braunschweig"
},
{
"operation": "Now on",
"time": "129.249",
"value": "Eintracht Braunschweig"
},
{
"operation": "search",
"time": "141.062",
"value": "2009-10 Eintracht Braunschweig"
},
{
"operation": "Now on",
"time": "142.308",
"value": "undefined"
},
{
"operation": "search",
"time": "156.544",
"value": "2009-10 Werder Bremen II season"
},
{
"operation": "Now on",
"time": "157.795",
"value": "2009–10 SV Werder Bremen II season"
},
{
"operation": "hyperlink",
"time": "159.172",
"value": "1.1.1 July and August 2009–10 SV Werder Bremen II season_section_2"
},
{
"operation": "Highlighting",
"time": "163.418",
"value": "2009–10 SV Werder Bremen II season_sentence_5"
},
{
"operation": "search",
"time": "185.48",
"value": "2009-10 Borussia Dortmund II season"
},
{
"operation": "Now on",
"time": "186.607",
"value": "2009–10 Borussia Dortmund II season"
},
{
"operation": "Highlighting",
"time": "207.309",
"value": "2009–10 Borussia Dortmund II season_sentence_5"
},
{
"operation": "Highlighting",
"time": "208.309",
"value": "2009–10 Borussia Dortmund II season_sentence_7"
},
{
"operation": "search",
"time": "220.848",
"value": "2009-10 Werder Bremen II season"
},
{
"operation": "Now on",
"time": "221.634",
"value": "2009-10 Werder Bremen II season"
},
{
"operation": "Highlighting",
"time": "229.333",
"value": "2009–10 SV Werder Bremen II season_sentence_5"
},
{
"operation": "finish",
"time": "259.103",
"value": "finish"
}
]
}
|
43987
|
Ana Uribe (born in Medellín, the second-largest city in Colombia) graduated in 1978 and then spent time in Brazil, later painting murals, including "Sunflowers, a Tribute to Aimee Willard."
|
{
"evidence": [
"Ana Uribe_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:\nAna Uribe (born in Medellín, the second-largest city in Colombia) graduated in 1978 and then spent time in Brazil, later painting murals, including \"Sunflowers, a Tribute to Aimee Willard.\"\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Ana Uribe_sentence_5"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 6999,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.018",
"value": "?search="
},
{
"operation": "search",
"time": "26.726",
"value": "Ana Uribe"
},
{
"operation": "Now on",
"time": "27.55",
"value": "Ana Uribe"
},
{
"operation": "Highlighting",
"time": "34.276",
"value": "Ana Uribe_sentence_2"
},
{
"operation": "hyperlink",
"time": "38.858",
"value": "Medellín, Colombia"
},
{
"operation": "Now on",
"time": "40.726",
"value": "Medellín"
},
{
"operation": "Highlighting",
"time": "44.241",
"value": "Medellín_sentence_1"
},
{
"operation": "Highlighting",
"time": "44.243",
"value": "Medellín_title"
},
{
"operation": "back-button-clicked",
"time": "47.129",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "47.409",
"value": "Ana Uribe"
},
{
"operation": "Highlighting",
"time": "50.945",
"value": "Ana Uribe_sentence_5"
},
{
"operation": "Highlighting",
"time": "55.353",
"value": "Ana Uribe_sentence_5"
},
{
"operation": "finish",
"time": "79.129",
"value": "finish"
}
]
}
|
exfever_3795
|
The Caracazo is the deadliest historical event that happened in Venezuela.
|
{
"evidence": [
"Caracazo",
"Guarenas"
],
"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 Caracazo is the deadliest historical event that happened in Venezuela.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Caracazo",
"Guarenas"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7000,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
66576
|
Thomas Wedge (a rugby union player born neither in St.Ives nor in Cornwall) and Maffer Davey are halfbacks playing for Great Britain (Cornwall) in the Rugby union at the 1908 Summer Olympics.
|
{
"evidence": [
"Thomas Wedge (rugby union)_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:\nThomas Wedge (a rugby union player born neither in St.Ives nor in Cornwall) and Maffer Davey are halfbacks playing for Great Britain (Cornwall) in the Rugby union at the 1908 Summer Olympics.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Thomas Wedge (rugby union)_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7001,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.97",
"value": "?search="
},
{
"operation": "search",
"time": "16.823",
"value": "Thomas Wedge (rugby union)"
},
{
"operation": "Now on",
"time": "17.661",
"value": "Thomas Wedge (rugby union)"
},
{
"operation": "Highlighting",
"time": "20.314",
"value": "Thomas Wedge (rugby union)_sentence_0"
},
{
"operation": "search",
"time": "33.42",
"value": "Maffer Davey"
},
{
"operation": "Now on",
"time": "34.305",
"value": "James Davey"
},
{
"operation": "Page search",
"time": "63.042",
"value": "halfback"
},
{
"operation": "back-button-clicked",
"time": "66.34",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "66.448",
"value": "Thomas Wedge (rugby union)"
},
{
"operation": "finish",
"time": "93.273",
"value": "finish"
}
]
}
|
exfever_3462
|
South Park is an American animated sitcom starring a fictional character voiced by Matt Stone.
|
{
"evidence": [
"South Park",
"Kyle Broflovski"
],
"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:\nSouth Park is an American animated sitcom starring a fictional character voiced by Matt Stone.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"South Park",
"Kyle Broflovski"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7002,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4075
|
Hannah Dakota Fanning rose to prominence at the age of seven for her performance as Lucy Dawson in a 2001 American drama film, in which an American actress known for numerous roles across stage and screen appeared in supporting roles.
|
{
"evidence": [
"Dakota Fanning",
"I Am Sam",
"Loretta Devine"
],
"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:\nHannah Dakota Fanning rose to prominence at the age of seven for her performance as Lucy Dawson in a 2001 American drama film, in which an American actress known for numerous roles across stage and screen appeared in supporting roles.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Dakota Fanning",
"I Am Sam",
"Loretta Devine"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7003,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4743
|
Cinderella is a romantic fantasy film directed by a British actor and filmmaker who won an Academy Award, four award for an independent charity that supports, develops, and promotes the art forms of the moving image.
|
{
"evidence": [
"Cinderella (2015 Disney film)",
"British Academy of Film and Television Arts"
],
"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:\nCinderella is a romantic fantasy film directed by a British actor and filmmaker who won an Academy Award, four award for an independent charity that supports, develops, and promotes the art forms of the moving image.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Cinderella (2015 Disney film)",
"British Academy of Film and Television Arts"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7004,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
35504
|
Amy and the Orphans directed by Scott Ellis, opened Off-Broadway (a seating capacity between 100 and 499) and is about three adult siblings
|
{
"evidence": [
"Amy and the Orphans_sentence_4",
"Amy and the Orphans_sentence_1",
"Off-Broadway_sentence_2",
"Amy and the Orphans_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nAmy and the Orphans directed by Scott Ellis, opened Off-Broadway (a seating capacity between 100 and 499) and is about three adult siblings\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Amy and the Orphans_sentence_4",
"Amy and the Orphans_sentence_1",
"Off-Broadway_sentence_2",
"Amy and the Orphans_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7005,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.687",
"value": "?search="
},
{
"operation": "search",
"time": "28.955",
"value": "Amy and the Orphans"
},
{
"operation": "Now on",
"time": "29.737",
"value": "Amy and the Orphans"
},
{
"operation": "Page search",
"time": "44.894",
"value": "direct"
},
{
"operation": "Highlighting",
"time": "47.645",
"value": "Amy and the Orphans_sentence_4"
},
{
"operation": "Highlighting",
"time": "51.382",
"value": "Amy and the Orphans_sentence_1"
},
{
"operation": "hyperlink",
"time": "56.23",
"value": "Off-Broadway"
},
{
"operation": "Now on",
"time": "56.955",
"value": "Off-Broadway"
},
{
"operation": "Highlighting",
"time": "61.857",
"value": "Off-Broadway_sentence_2"
},
{
"operation": "back-button-clicked",
"time": "64.919",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "65.121",
"value": "Amy and the Orphans"
},
{
"operation": "Highlighting",
"time": "68.975",
"value": "Amy and the Orphans_sentence_2"
},
{
"operation": "finish",
"time": "248.431",
"value": "finish"
}
]
}
|
39743
|
Allahabad is likely to get a ropeway facility over Yamuna and the tourism department has forwarded a proposal to the state government for approval.
|
{
"evidence": [
"Saraswati Ghat_sentence_8",
"Saraswati Ghat_sentence_9"
],
"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:\nAllahabad is likely to get a ropeway facility over Yamuna and the tourism department has forwarded a proposal to the state government for approval.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Saraswati Ghat_sentence_8",
"Saraswati Ghat_sentence_9"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7006,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.887",
"value": "?search="
},
{
"operation": "search",
"time": "18.632",
"value": "\"tourism department has forwarded a proposal to the state government \""
},
{
"operation": "Now on",
"time": "19.793",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "21.643",
"value": "Saraswati Ghat"
},
{
"operation": "Now on",
"time": "22.303",
"value": "Saraswati Ghat"
},
{
"operation": "Page search",
"time": "29.602",
"value": "allahabad"
},
{
"operation": "Highlighting",
"time": "37.279",
"value": "Saraswati Ghat_sentence_8"
},
{
"operation": "Highlighting",
"time": "41.231",
"value": "Saraswati Ghat_sentence_9"
},
{
"operation": "finish",
"time": "60.653",
"value": "finish"
}
]
}
|
exfever_3556
|
James Franco has a wide range of accomplishments, the actor will add a doctorate in English from Yale University to his list all while working on film and writing projects.
|
{
"evidence": [
"Oz the Great and Powerful",
"The Wizard of Oz (1939 film)"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJames Franco has a wide range of accomplishments, the actor will add a doctorate in English from Yale University to his list all while working on film and writing projects.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Oz the Great and Powerful",
"The Wizard of Oz (1939 film)"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7007,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8955
|
Nicole Scherzinger is an American singer-songwriter raised in the nominal seat of a county located in the north central portion of the United States, the Commonwealth of Kentucky.
|
{
"evidence": [
"Nicole Scherzinger",
"Louisville, Kentucky",
"Jefferson County, Kentucky"
],
"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:\nNicole Scherzinger is an American singer-songwriter raised in the nominal seat of a county located in the north central portion of the United States, the Commonwealth of Kentucky.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Nicole Scherzinger",
"Louisville, Kentucky",
"Jefferson County, Kentucky"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7008,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
58993
|
The International Quorum of Motion Picture Producers is comprised more than 80 members in December 2010 and includes various film production companies.
|
{
"evidence": [
"International Quorum of Motion Picture Producers_sentence_6",
"International Quorum of Motion Picture Producers_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe International Quorum of Motion Picture Producers is comprised more than 80 members in December 2010 and includes various film production companies.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"International Quorum of Motion Picture Producers_sentence_6",
"International Quorum of Motion Picture Producers_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7009,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.161",
"value": "?search="
},
{
"operation": "search",
"time": "53.556",
"value": "International Quorum of Motion Picture Producers"
},
{
"operation": "Now on",
"time": "54.276",
"value": "International Quorum of Motion Picture Producers"
},
{
"operation": "Highlighting",
"time": "138.447",
"value": "International Quorum of Motion Picture Producers_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "147.91",
"value": "International Quorum of Motion Picture Producers_sentence_2"
},
{
"operation": "Highlighting",
"time": "148.914",
"value": "International Quorum of Motion Picture Producers_sentence_6"
},
{
"operation": "Page search",
"time": "169.388",
"value": "film p"
},
{
"operation": "Highlighting",
"time": "175.91",
"value": "International Quorum of Motion Picture Producers_sentence_2"
},
{
"operation": "finish",
"time": "188.315",
"value": "finish"
}
]
}
|
92061
|
Barangaroo ferry wharf was the only wharf to service Darling Harbour until Pyrmont Bay (located on the western side of Darling Harbour) was opened in the early 2000s.
|
{
"evidence": [
"Barangaroo ferry wharf_sentence_9",
"Pyrmont Bay ferry wharf_sentence_0"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBarangaroo ferry wharf was the only wharf to service Darling Harbour until Pyrmont Bay (located on the western side of Darling Harbour) was opened in the early 2000s.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Barangaroo ferry wharf_sentence_9",
"Pyrmont Bay ferry wharf_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7010,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.454",
"value": "?search="
},
{
"operation": "search",
"time": "15.718",
"value": "Barangaroo ferry wharf"
},
{
"operation": "Now on",
"time": "16.681",
"value": "Barangaroo ferry wharf"
},
{
"operation": "Highlighting",
"time": "36.492",
"value": "Barangaroo ferry wharf_sentence_9"
},
{
"operation": "hyperlink",
"time": "41.691",
"value": "Pyrmont Bay"
},
{
"operation": "Now on",
"time": "42.244",
"value": "Pyrmont Bay ferry wharf"
},
{
"operation": "Highlighting",
"time": "43.356",
"value": "Pyrmont Bay ferry wharf_sentence_0"
},
{
"operation": "finish",
"time": "47.299",
"value": "finish"
}
]
}
|
exfever_3742
|
Gwen Stefani is the primary songwriter for the band No Doubt, whose singles include a song from their third studio album, Tragic Kingdom, released as the album's second single on November 19, 1995.
|
{
"evidence": [
"Gwen Stefani",
"Spiderwebs (song)"
],
"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:\nGwen Stefani is the primary songwriter for the band No Doubt, whose singles include a song from their third studio album, Tragic Kingdom, released as the album's second single on November 19, 1995.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Gwen Stefani",
"Spiderwebs (song)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7011,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
47043
|
The Oriental hornet (Vespa orientalis, Linnaeus, 1761) is a social insect of the family Vespidae that is found in Southwest Asia, Northeast Africa, the island of Madagascar, the Middle East, Central Asia, and parts of Southern Europe.
|
{
"evidence": [
"Oriental hornet_sentence_0",
"Oriental hornet_sentence_1",
"Carl Linnaeus_sentence_238",
"Carl Linnaeus_sentence_6",
"Carl Linnaeus_sentence_4"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Oriental hornet (Vespa orientalis, Linnaeus, 1761) is a social insect of the family Vespidae that is found in Southwest Asia, Northeast Africa, the island of Madagascar, the Middle East, Central Asia, and parts of Southern Europe.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Oriental hornet_sentence_0",
"Oriental hornet_sentence_1",
"Carl Linnaeus_sentence_238",
"Carl Linnaeus_sentence_6",
"Carl Linnaeus_sentence_4"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7012,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.3",
"value": "?search="
},
{
"operation": "search",
"time": "19.678",
"value": "Contains...The Oriental hornet"
},
{
"operation": "Now on",
"time": "21.325",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "23.794",
"value": "Oriental hornet"
},
{
"operation": "Now on",
"time": "24.417",
"value": "Oriental hornet"
},
{
"operation": "Highlighting",
"time": "28.462",
"value": "Oriental hornet_sentence_0"
},
{
"operation": "Highlighting",
"time": "47.997",
"value": "Oriental hornet_sentence_1"
},
{
"operation": "search",
"time": "75.163",
"value": "Vespa orientalis"
},
{
"operation": "Now on",
"time": "76.473",
"value": "Oriental hornet"
},
{
"operation": "search",
"time": "197.604",
"value": "Linnaeus"
},
{
"operation": "Now on",
"time": "198.823",
"value": "Carl Linnaeus"
},
{
"operation": "hyperlink",
"time": "206.2",
"value": "Linnaeus (disambiguation)"
},
{
"operation": "Now on",
"time": "207.097",
"value": "Linnaeus (disambiguation)"
},
{
"operation": "hyperlink",
"time": "221.278",
"value": "Carl Linnaeus"
},
{
"operation": "Now on",
"time": "222.509",
"value": "Carl Linnaeus"
},
{
"operation": "Highlighting",
"time": "273.189",
"value": "Carl Linnaeus_sentence_238"
},
{
"operation": "Highlighting",
"time": "288.429",
"value": "Carl Linnaeus_sentence_6"
},
{
"operation": "Highlighting",
"time": "300.922",
"value": "Carl Linnaeus_sentence_4"
},
{
"operation": "hyperlink",
"time": "301.715",
"value": "binomial nomenclature"
},
{
"operation": "Now on",
"time": "302.736",
"value": "Binomial nomenclature"
},
{
"operation": "finish",
"time": "317.491",
"value": "finish"
}
]
}
|
50841
|
Minkhaung Medaw had seven siblings.
|
{
"evidence": [
"Minkhaung Medaw_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:\nMinkhaung Medaw had seven siblings.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Minkhaung Medaw_sentence_6"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7013,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.304",
"value": "?search="
},
{
"operation": "search",
"time": "13.01",
"value": "Minkhaung Medaw"
},
{
"operation": "Now on",
"time": "14.014",
"value": "Minkhaung Medaw"
},
{
"operation": "Highlighting",
"time": "29.76",
"value": "Minkhaung Medaw_sentence_6"
},
{
"operation": "finish",
"time": "64.051",
"value": "finish"
}
]
}
|
63244
|
Bungee balls became a fad in February 1991 and were recalled in October 2003 for containing synthetic liquid.
|
{
"evidence": [
"Bungee ball_sentence_3",
"Bungee ball_sentence_6",
"Bungee ball_sentence_5",
"Bungee ball_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBungee balls became a fad in February 1991 and were recalled in October 2003 for containing synthetic liquid.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Bungee ball_sentence_3",
"Bungee ball_sentence_6",
"Bungee ball_sentence_5",
"Bungee ball_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7014,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.303",
"value": "?search="
},
{
"operation": "search",
"time": "10.453",
"value": "Contains...Bungee balls"
},
{
"operation": "Now on",
"time": "12.021",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "14.577",
"value": "Bungee ball"
},
{
"operation": "Now on",
"time": "15.292",
"value": "Bungee ball"
},
{
"operation": "Highlighting",
"time": "24.961",
"value": "Bungee ball_sentence_3"
},
{
"operation": "Highlighting",
"time": "32.585",
"value": "Bungee ball_sentence_6"
},
{
"operation": "Highlighting",
"time": "36.688",
"value": "Bungee ball_sentence_5"
},
{
"operation": "Highlighting",
"time": "44.888",
"value": "Bungee ball_sentence_2"
},
{
"operation": "finish",
"time": "51.609",
"value": "finish"
}
]
}
|
exfever_6303
|
Jared Leto is an American musician who had supporting roles in an independent film genre that seeks to elicit fear or disgust in its audience for entertainment purposes.
|
{
"evidence": [
"Jared Leto",
"American Psycho (film)",
"Horror 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:\nJared Leto is an American musician who had supporting roles in an independent film genre that seeks to elicit fear or disgust in its audience for entertainment purposes.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Jared Leto",
"American Psycho (film)",
"Horror film"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7015,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3119
|
The X Factor is an American television music competition to find new singing talent, presented by Konnie Huq on ITV2, who took over from an English television presenter, model, and author.
|
{
"evidence": [
"The X Factor (UK series 7)",
"Holly Willoughby"
],
"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 X Factor is an American television music competition to find new singing talent, presented by Konnie Huq on ITV2, who took over from an English television presenter, model, and author.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The X Factor (UK series 7)",
"Holly Willoughby"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7016,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
57996
|
Oded Galor founded the field of unilateral space theory, which suggests that during most of human existence, technological progress was counterbalanced by population growth and living standards were near subsistence across time and space.
|
{
"evidence": [
"Unified growth theory_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:\nOded Galor founded the field of unilateral space theory, which suggests that during most of human existence, technological progress was counterbalanced by population growth and living standards were near subsistence across time and space.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Unified growth theory_sentence_7"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7017,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.729",
"value": "?search="
},
{
"operation": "search",
"time": "103.251",
"value": "Oded Galor"
},
{
"operation": "Now on",
"time": "104.965",
"value": "Oded Galor"
},
{
"operation": "Highlighting",
"time": "190.264",
"value": "Oded Galor_sentence_9"
},
{
"operation": "Highlighting",
"time": "190.264",
"value": "Oded Galor_title"
},
{
"operation": "Highlighting",
"time": "190.265",
"value": "Oded Galor_section_1"
},
{
"operation": "hyperlink",
"time": "198.406",
"value": "unified growth theory"
},
{
"operation": "Now on",
"time": "200.043",
"value": "Unified growth theory"
},
{
"operation": "Highlighting",
"time": "239.254",
"value": "Unified growth theory_sentence_7"
},
{
"operation": "finish",
"time": "273.445",
"value": "finish"
}
]
}
|
exfever_3087
|
The Man with the Iron Fists earned $20.5 million at the box office during Thursday night preview screenings.
|
{
"evidence": [
"The Man with the Iron Fists",
"Dave Bautista"
],
"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 Man with the Iron Fists earned $20.5 million at the box office during Thursday night preview screenings.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Man with the Iron Fists",
"Dave Bautista"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7018,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
44927
|
He was arrested in France on October 8, 2009, after police allegedly intercepted emails between himself and Al-Qaida in Islamic Maghreb, an Algerian-based terrorist organisation.
|
{
"evidence": [
"Adlène Hicheur_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:\nHe was arrested in France on October 8, 2009, after police allegedly intercepted emails between himself and Al-Qaida in Islamic Maghreb, an Algerian-based terrorist organisation.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Adlène Hicheur_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7019,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.27",
"value": "?search="
},
{
"operation": "search",
"time": "25.214",
"value": "Adlène Hicheur"
},
{
"operation": "Now on",
"time": "26.584",
"value": "Adlène Hicheur"
},
{
"operation": "Highlighting",
"time": "39.217",
"value": "Adlène Hicheur_sentence_5"
},
{
"operation": "finish",
"time": "49.459",
"value": "finish"
}
]
}
|
exfever_3738
|
The Coen brothers are American filmmakers whose most acclaimed works include a 1987 American crime comedy film.
|
{
"evidence": [
"Coen brothers",
"Raising Arizona"
],
"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 Coen brothers are American filmmakers whose most acclaimed works include a 1987 American crime comedy film.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Coen brothers",
"Raising Arizona"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7020,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4694
|
Meek Mill is an Indian rapper who signed with a record label imprint founded by Miami-based American rapper Rick Ross, which albums are distributed by an American record label founded in October 1947 by Ahmet Ertegun and Herb Abramson.
|
{
"evidence": [
"Meek Mill",
"Maybach Music Group",
"Atlantic Records"
],
"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:\nMeek Mill is an Indian rapper who signed with a record label imprint founded by Miami-based American rapper Rick Ross, which albums are distributed by an American record label founded in October 1947 by Ahmet Ertegun and Herb Abramson.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Meek Mill",
"Maybach Music Group",
"Atlantic Records"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7021,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7576
|
Zooey Claire Deschanel is an American actress known for her deadpan role in a 2003 American teen comedy film that stars an American actress and producer best known for starring Faith in the supernatural drama series Buffy the Vampire Slayer.
|
{
"evidence": [
"Zooey Deschanel",
"The New Guy",
"Eliza Dushku"
],
"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:\nZooey Claire Deschanel is an American actress known for her deadpan role in a 2003 American teen comedy film that stars an American actress and producer best known for starring Faith in the supernatural drama series Buffy the Vampire Slayer.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Zooey Deschanel",
"The New Guy",
"Eliza Dushku"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7022,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
11695
|
The two separate qualifying tournaments in the new system is the Champions Path, for clubs which won their domestic league and did not automatically qualify for the group stage, and the Non-Champions Path, for clubs which did not win their domestic league and did not automatically qualify for the group stage.
|
{
"evidence": [
"UEFA Women's Champions League_sentence_34"
],
"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 two separate qualifying tournaments in the new system is the Champions Path, for clubs which won their domestic league and did not automatically qualify for the group stage, and the Non-Champions Path, for clubs which did not win their domestic league and did not automatically qualify for the group stage.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"UEFA Women's Champions League_sentence_34"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7023,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.943",
"value": "?search="
},
{
"operation": "search",
"time": "26.802",
"value": "Champions Path"
},
{
"operation": "Now on",
"time": "34.826",
"value": "undefined"
},
{
"operation": "search",
"time": "77.715",
"value": "non-champions path"
},
{
"operation": "Now on",
"time": "80.512",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "86.026",
"value": "2010–11 UEFA Champions League group stage"
},
{
"operation": "Now on",
"time": "91.164",
"value": "2010–11 UEFA Champions League group stage"
},
{
"operation": "Highlighting",
"time": "111.55",
"value": "2010–11 UEFA Champions League group stage_sentence_0"
},
{
"operation": "Highlighting",
"time": "111.551",
"value": "2010–11 UEFA Champions League group stage_title"
},
{
"operation": "search",
"time": "118.314",
"value": "UEFA Champions League"
},
{
"operation": "Now on",
"time": "129.98",
"value": "UEFA Champions League"
},
{
"operation": "Page search",
"time": "155.03",
"value": "champions path"
},
{
"operation": "Page search",
"time": "182.198",
"value": "non-champions path"
},
{
"operation": "search",
"time": "287.626",
"value": "non-champions path"
},
{
"operation": "Now on",
"time": "292.797",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "305.218",
"value": "UEFA Women's Champions League"
},
{
"operation": "Now on",
"time": "311.099",
"value": "UEFA Women's Champions League"
},
{
"operation": "Page search",
"time": "328.98",
"value": "champions path"
},
{
"operation": "Page search",
"time": "347.411",
"value": "non-champions path"
},
{
"operation": "Highlighting",
"time": "380.067",
"value": "UEFA Women's Champions League_sentence_34"
},
{
"operation": "Now on",
"time": "468.181",
"value": "UEFA Women's Champions League"
},
{
"operation": "finish",
"time": "484.266",
"value": "finish"
}
]
}
|
exfever_8943
|
The Gadsden flag is a historical American flag named an American politician who was the principal leader of the South Carolina Patriot movement during the American Revolution.
|
{
"evidence": [
"Gadsden flag",
"Christopher Gadsden"
],
"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 Gadsden flag is a historical American flag named an American politician who was the principal leader of the South Carolina Patriot movement during the American Revolution.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Gadsden flag",
"Christopher Gadsden"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7024,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
83709
|
Despite cold War tensions Mikhail Chumakov, a Soviet microbiologist and virologist, traveled to the U.S. to study the Salk polio vaccine.
|
{
"evidence": [
"Mikhail Chumakov_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:\nDespite cold War tensions Mikhail Chumakov, a Soviet microbiologist and virologist, traveled to the U.S. to study the Salk polio vaccine.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Mikhail Chumakov_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7025,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.594",
"value": "undefined"
},
{
"operation": "search",
"time": "5.606",
"value": "Mikhail Chumakov,"
},
{
"operation": "Now on",
"time": "6.338",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "14.324",
"value": "Mikhail Chumakov"
},
{
"operation": "Now on",
"time": "14.847",
"value": "Mikhail Chumakov"
},
{
"operation": "Highlighting",
"time": "22.773",
"value": "Mikhail Chumakov_sentence_0"
},
{
"operation": "Page search",
"time": "31.741",
"value": "united states"
},
{
"operation": "Page search",
"time": "56.787",
"value": "travel"
},
{
"operation": "Page search",
"time": "62.235",
"value": "salk"
},
{
"operation": "Page search",
"time": "82.46",
"value": "polio"
},
{
"operation": "hyperlink",
"time": "132.534",
"value": "Polio Vaccine"
},
{
"operation": "Now on",
"time": "133.947",
"value": "Polio vaccine"
},
{
"operation": "Page search",
"time": "137.868",
"value": "Chumakov"
},
{
"operation": "search",
"time": "157.554",
"value": "Salk polio vaccine"
},
{
"operation": "Now on",
"time": "158.726",
"value": "Polio vaccine"
},
{
"operation": "Page search",
"time": "164.141",
"value": "Salk"
},
{
"operation": "hyperlink",
"time": "175.137",
"value": "4 History Polio vaccine_section_10"
},
{
"operation": "Page search",
"time": "217.225",
"value": "Chumakov"
},
{
"operation": "finish",
"time": "248.821",
"value": "finish"
}
]
}
|
30071
|
After Maria did not return home one night, Borowski began to suspect that she had been arrested; but rather than staying away from any of their usual meeting places, though, he walked straight into the trap that was set by the Gestapo agents in the apartment of his and Maria's close friend.
|
{
"evidence": [
"Tadeusz Borowski_sentence_21",
"Tadeusz Borowski_sentence_22"
],
"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:\nAfter Maria did not return home one night, Borowski began to suspect that she had been arrested; but rather than staying away from any of their usual meeting places, though, he walked straight into the trap that was set by the Gestapo agents in the apartment of his and Maria's close friend.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Tadeusz Borowski_sentence_21",
"Tadeusz Borowski_sentence_22"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7026,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.507",
"value": "?search="
},
{
"operation": "search",
"time": "103.413",
"value": "Borowski"
},
{
"operation": "Now on",
"time": "104.21",
"value": "Borowski"
},
{
"operation": "search",
"time": "225.05",
"value": "Gestapo"
},
{
"operation": "Now on",
"time": "226.343",
"value": "Gestapo"
},
{
"operation": "search",
"time": "236.019",
"value": "After Maria did not return home one night, Borowski began to suspect that she had been arrested"
},
{
"operation": "Now on",
"time": "238.412",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "242.318",
"value": "After the war Tadeusz Borowski_section_2"
},
{
"operation": "Now on",
"time": "242.675",
"value": "Tadeusz Borowski#After the war Tadeusz Borowski section 2"
},
{
"operation": "Page search",
"time": "266.578",
"value": "After Maria did not return home one night, Borowski began to suspect that she had been arrested"
},
{
"operation": "Page search",
"time": "272.001",
"value": "Maria"
},
{
"operation": "Highlighting",
"time": "277.453",
"value": "Tadeusz Borowski_sentence_21"
},
{
"operation": "Highlighting",
"time": "285.249",
"value": "Tadeusz Borowski_sentence_22"
},
{
"operation": "finish",
"time": "343.29",
"value": "finish"
}
]
}
|
77491
|
Friedrich von Bodelschwingh, a German pastor, theologian and public health advocate, was the son of Reverend Friedrich Christian Carl von Bodelschwingh and is sometimes known as Friedrich von Bodelschwingh the Younger to distinguish him from his father.
|
{
"evidence": [
"Friedrich von Bodelschwingh_sentence_0",
"Friedrich von Bodelschwingh_sentence_3",
"Friedrich von Bodelschwingh_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:\nFriedrich von Bodelschwingh, a German pastor, theologian and public health advocate, was the son of Reverend Friedrich Christian Carl von Bodelschwingh and is sometimes known as Friedrich von Bodelschwingh the Younger to distinguish him from his father.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Friedrich von Bodelschwingh_sentence_0",
"Friedrich von Bodelschwingh_sentence_3",
"Friedrich von Bodelschwingh_sentence_4"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7027,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.361",
"value": "?search="
},
{
"operation": "search",
"time": "71.769",
"value": "Friedrich von Bodelschwingh"
},
{
"operation": "Now on",
"time": "72.928",
"value": "Friedrich von Bodelschwingh"
},
{
"operation": "Highlighting",
"time": "114.98",
"value": "Friedrich von Bodelschwingh_sentence_0"
},
{
"operation": "Highlighting",
"time": "214.566",
"value": "Friedrich von Bodelschwingh_sentence_3"
},
{
"operation": "Highlighting",
"time": "236.763",
"value": "Friedrich von Bodelschwingh_sentence_4"
},
{
"operation": "finish",
"time": "582.222",
"value": "finish"
}
]
}
|
exfever_1407
|
New Horizons is an interplanetary space probe that was launched as a part of a series of ocean exploration missions being conducted by NASA.
|
{
"evidence": [
"New Horizons",
"New Frontiers program"
],
"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:\nNew Horizons is an interplanetary space probe that was launched as a part of a series of ocean exploration missions being conducted by NASA.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"New Horizons",
"New Frontiers program"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7028,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2108
|
David Angell was an American screenwriter who died heading home from their vacation on Cape Cod aboard a domestic passenger flight that was hijacked by five al-Qaeda terrorists on September 11, 2005.
|
{
"evidence": [
"David Angell",
"American Airlines Flight 11"
],
"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:\nDavid Angell was an American screenwriter who died heading home from their vacation on Cape Cod aboard a domestic passenger flight that was hijacked by five al-Qaeda terrorists on September 11, 2005.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"David Angell",
"American Airlines Flight 11"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7029,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
20838
|
Living Marxism writers applauded the media for their portrayal of the civil wars in Rwanda in Bosnia, believing the Serb and Hutu forces committed genocide, later publishing an exhibition showing images of deaths in these conflicts.
|
{
"evidence": [
"Living Marxism_sentence_10"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nLiving Marxism writers applauded the media for their portrayal of the civil wars in Rwanda in Bosnia, believing the Serb and Hutu forces committed genocide, later publishing an exhibition showing images of deaths in these conflicts.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Living Marxism_sentence_10"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7030,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.644",
"value": "?search="
},
{
"operation": "search",
"time": "11.708",
"value": "Living Marxism"
},
{
"operation": "Now on",
"time": "12.655",
"value": "Living Marxism"
},
{
"operation": "Page search",
"time": "86.533",
"value": "rwanda"
},
{
"operation": "Highlighting",
"time": "104.487",
"value": "Living Marxism_sentence_10"
},
{
"operation": "finish",
"time": "191.701",
"value": "finish"
}
]
}
|
exfever_6323
|
Poland is the 8th richest country in the world.
|
{
"evidence": [
"Poland",
"Warsaw"
],
"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:\nPoland is the 8th richest country in the world.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Poland",
"Warsaw"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7031,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
11540
|
Erich von Holst(28 November 1908 – 26 May 1962) was a German behavioral physiologist and founder of the Max Planck Institute for Behavioral Physiology from Riga, Livonia who was remembered for his work concerning the processes of endogenous generation of stimuli and of central coordination as a basis of behavioral physiology.
|
{
"evidence": [
"Erich von Holst_sentence_0",
"Erich von Holst_sentence_1",
"Erich von Holst_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nErich von Holst(28 November 1908 – 26 May 1962) was a German behavioral physiologist and founder of the Max Planck Institute for Behavioral Physiology from Riga, Livonia who was remembered for his work concerning the processes of endogenous generation of stimuli and of central coordination as a basis of behavioral physiology.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Erich von Holst_sentence_0",
"Erich von Holst_sentence_1",
"Erich von Holst_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7032,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.498",
"value": "?search="
},
{
"operation": "search",
"time": "24.657",
"value": "Erich von Holst"
},
{
"operation": "Now on",
"time": "25.587",
"value": "Erich von Holst"
},
{
"operation": "Highlighting",
"time": "40.519",
"value": "Erich von Holst_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "50.666",
"value": "Erich von Holst_sentence_1"
},
{
"operation": "Highlighting",
"time": "51.177",
"value": "Erich von Holst_sentence_0"
},
{
"operation": "Highlighting",
"time": "51.881",
"value": "Erich von Holst_sentence_1"
},
{
"operation": "Highlighting",
"time": "76.207",
"value": "Erich von Holst_sentence_2"
},
{
"operation": "finish",
"time": "101.417",
"value": "finish"
}
]
}
|
32431
|
The community population of Markopoulo Oropou, a town located in Nothern Europe, increases steadily every ten years, from a population of 2,006 in 1981 to 2,789 in 2011.
|
{
"evidence": [
"Markopoulo Oropou_sentence_0",
"Greece_sentence_1"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe community population of Markopoulo Oropou, a town located in Nothern Europe, increases steadily every ten years, from a population of 2,006 in 1981 to 2,789 in 2011.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Markopoulo Oropou_sentence_0",
"Greece_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7033,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.546",
"value": "?search="
},
{
"operation": "search",
"time": "11.019",
"value": "Markopoulo Oropou"
},
{
"operation": "Now on",
"time": "11.819",
"value": "Markopoulo Oropou"
},
{
"operation": "Highlighting",
"time": "17.061",
"value": "Markopoulo Oropou_sentence_0"
},
{
"operation": "hyperlink",
"time": "33.763",
"value": "Greece"
},
{
"operation": "Now on",
"time": "34.573",
"value": "Greece"
},
{
"operation": "Highlighting",
"time": "43.679",
"value": "Greece_sentence_1"
},
{
"operation": "back-button-clicked",
"time": "67.066",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "67.124",
"value": "Markopoulo Oropou"
},
{
"operation": "Highlighting",
"time": "76.113",
"value": "Markopoulo Oropou_cell_0_3_1"
},
{
"operation": "Highlighting",
"time": "76.118",
"value": "Markopoulo Oropou_header_cell_0_3_0"
},
{
"operation": "Highlighting",
"time": "76.118",
"value": "Markopoulo Oropou_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "76.747",
"value": "Markopoulo Oropou_cell_0_3_1"
},
{
"operation": "Highlighting deleted",
"time": "76.753",
"value": "Markopoulo Oropou_header_cell_0_3_0"
},
{
"operation": "Highlighting deleted",
"time": "76.754",
"value": "Markopoulo Oropou_header_cell_0_0_0"
},
{
"operation": "hyperlink",
"time": "77.317",
"value": "East Attica"
},
{
"operation": "Now on",
"time": "77.771",
"value": "East Attica"
},
{
"operation": "finish",
"time": "87.691",
"value": "finish"
}
]
}
|
exfever_1849
|
Ann Romney is a philanthropist who was also the First Lady of Massachusetts, while her husband served as the head of the state cabinet and the secretary of the commonwealth's military forces.
|
{
"evidence": [
"Ann Romney",
"Governor of Massachusetts"
],
"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:\nAnn Romney is a philanthropist who was also the First Lady of Massachusetts, while her husband served as the head of the state cabinet and the secretary of the commonwealth's military forces.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Ann Romney",
"Governor of Massachusetts"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7034,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_763
|
Baylor University is the eighth private research university in Texas.
|
{
"evidence": [
"Baylor University",
"Brazos River"
],
"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:\nBaylor University is the eighth private research university in Texas.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Baylor University",
"Brazos River"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7035,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
94769
|
Candle Night 2017 was a concert tour by Japanese singer Misia, who set up a poll through her official mobile app to allow fans to vote for which songs she should perform.
|
{
"evidence": [
"Misia Candle Night 2017_sentence_0",
"Misia Candle Night 2017_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:\nCandle Night 2017 was a concert tour by Japanese singer Misia, who set up a poll through her official mobile app to allow fans to vote for which songs she should perform.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Misia Candle Night 2017_sentence_0",
"Misia Candle Night 2017_sentence_8"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7036,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.649",
"value": "?search="
},
{
"operation": "search",
"time": "21.535",
"value": "Misia"
},
{
"operation": "Now on",
"time": "22.776",
"value": "Misia"
},
{
"operation": "Page search",
"time": "46.154",
"value": "concert"
},
{
"operation": "Page search",
"time": "75.618",
"value": "candle night"
},
{
"operation": "hyperlink",
"time": "91.977",
"value": "Misia discography"
},
{
"operation": "Now on",
"time": "93.061",
"value": "Misia discography"
},
{
"operation": "hyperlink",
"time": "108.967",
"value": "5 Live album Misia discography_section_4"
},
{
"operation": "search",
"time": "128.805",
"value": "candle night 2017"
},
{
"operation": "Now on",
"time": "131.114",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "134.698",
"value": "Misia Candle Night 2017"
},
{
"operation": "Now on",
"time": "135.847",
"value": "Misia Candle Night 2017"
},
{
"operation": "Highlighting",
"time": "141.351",
"value": "Misia Candle Night 2017_sentence_0"
},
{
"operation": "Page search",
"time": "157.943",
"value": "poll"
},
{
"operation": "Highlighting",
"time": "180.572",
"value": "Misia Candle Night 2017_sentence_8"
},
{
"operation": "finish",
"time": "224.089",
"value": "finish"
}
]
}
|
60323
|
Werner Veigel (9 November 1928 – 2 May 1995) was a Dutch-born German journalist and news presenter who is also known as a newsreader.
|
{
"evidence": [
"Werner Veigel_sentence_0",
"News presenter_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:\nWerner Veigel (9 November 1928 – 2 May 1995) was a Dutch-born German journalist and news presenter who is also known as a newsreader.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Werner Veigel_sentence_0",
"News presenter_sentence_4"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7037,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.332",
"value": "?search="
},
{
"operation": "search",
"time": "19.471",
"value": "Werner Veigel"
},
{
"operation": "Now on",
"time": "20.376",
"value": "Werner Veigel"
},
{
"operation": "Highlighting",
"time": "44.983",
"value": "Werner Veigel_sentence_0"
},
{
"operation": "hyperlink",
"time": "67.943",
"value": "news presenter"
},
{
"operation": "Now on",
"time": "68.758",
"value": "News presenter"
},
{
"operation": "Highlighting",
"time": "74.12",
"value": "News presenter_sentence_4"
},
{
"operation": "finish",
"time": "80.083",
"value": "finish"
}
]
}
|
38042
|
Kaye Abad starred in numerous teen-oriented films and TV series and is most remembered in Tabing Ilog with her screen partner John Lloyd Cruz.
|
{
"evidence": [
"Kaye Abad_sentence_4",
"Kaye Abad_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:\nKaye Abad starred in numerous teen-oriented films and TV series and is most remembered in Tabing Ilog with her screen partner John Lloyd Cruz.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Kaye Abad_sentence_4",
"Kaye Abad_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7038,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.48",
"value": "undefined"
},
{
"operation": "search",
"time": "23.344",
"value": "Kaye Abad"
},
{
"operation": "Now on",
"time": "24.05",
"value": "Kaye Abad"
},
{
"operation": "Page search",
"time": "55.773",
"value": "teen"
},
{
"operation": "Highlighting",
"time": "59.772",
"value": "Kaye Abad_sentence_4"
},
{
"operation": "Highlighting",
"time": "78.273",
"value": "Kaye Abad_sentence_5"
},
{
"operation": "Page search",
"time": "97.376",
"value": "john"
},
{
"operation": "hyperlink",
"time": "106.393",
"value": "Tabing Ilog"
},
{
"operation": "Now on",
"time": "106.958",
"value": "Tabing Ilog"
},
{
"operation": "Highlighting",
"time": "115.454",
"value": "Tabing Ilog_sentence_2Tabing Ilog_sentence_3"
},
{
"operation": "Highlighting",
"time": "115.455",
"value": "Tabing Ilog_title"
},
{
"operation": "back-button-clicked",
"time": "141.154",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "141.323",
"value": "Kaye Abad"
},
{
"operation": "hyperlink",
"time": "166.193",
"value": "Tabing Ilog"
},
{
"operation": "Now on",
"time": "166.566",
"value": "Tabing Ilog"
},
{
"operation": "back-button-clicked",
"time": "171.896",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "172.092",
"value": "Kaye Abad"
},
{
"operation": "Page search",
"time": "217.399",
"value": "john"
},
{
"operation": "hyperlink",
"time": "235.669",
"value": "Tabing Ilog"
},
{
"operation": "Now on",
"time": "236.034",
"value": "Tabing Ilog"
},
{
"operation": "Highlighting deleted",
"time": "238.268",
"value": "Tabing Ilog_sentence_2Tabing Ilog_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "238.27",
"value": "Tabing Ilog_title"
},
{
"operation": "Highlighting",
"time": "242.705",
"value": "Tabing Ilog_item_1_2"
},
{
"operation": "Highlighting",
"time": "242.707",
"value": "Tabing Ilog_title"
},
{
"operation": "Highlighting",
"time": "242.707",
"value": "Tabing Ilog_section_0"
},
{
"operation": "Highlighting",
"time": "242.708",
"value": "Tabing Ilog_section_1"
},
{
"operation": "Highlighting deleted",
"time": "261.689",
"value": "Tabing Ilog_item_1_2"
},
{
"operation": "Highlighting deleted",
"time": "261.691",
"value": "Tabing Ilog_title"
},
{
"operation": "Highlighting deleted",
"time": "261.693",
"value": "Tabing Ilog_section_0"
},
{
"operation": "Highlighting deleted",
"time": "261.695",
"value": "Tabing Ilog_section_1"
},
{
"operation": "back-button-clicked",
"time": "262.791",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "262.973",
"value": "Kaye Abad"
},
{
"operation": "finish",
"time": "294.395",
"value": "finish"
}
]
}
|
73333
|
The 5th European Athletics Championships were held at Stadion Neufeld from 25–29 August 1954 in the Swiss capital Bern.
|
{
"evidence": [
"1954 European Athletics Championships_sentence_0"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe 5th European Athletics Championships were held at Stadion Neufeld from 25–29 August 1954 in the Swiss capital Bern.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"1954 European Athletics Championships_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7039,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.478",
"value": "?search="
},
{
"operation": "search",
"time": "14.927",
"value": "Contains...5th European Athletics Championship"
},
{
"operation": "Now on",
"time": "18.4",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "22.355",
"value": "1954 European Athletics Championships"
},
{
"operation": "Now on",
"time": "24.591",
"value": "1954 European Athletics Championships"
},
{
"operation": "Highlighting",
"time": "26.654",
"value": "1954 European Athletics Championships_sentence_0"
},
{
"operation": "finish",
"time": "53.125",
"value": "finish"
}
]
}
|
75916
|
Sri Lankan diplomat Hamilton Shirley Amerasinghe did not hold any UN positions.
|
{
"evidence": [
"Hamilton Shirley Amerasinghe_sentence_4",
"Hamilton Shirley Amerasinghe_sentence_5",
"Hamilton Shirley Amerasinghe_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:\nSri Lankan diplomat Hamilton Shirley Amerasinghe did not hold any UN positions.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Hamilton Shirley Amerasinghe_sentence_4",
"Hamilton Shirley Amerasinghe_sentence_5",
"Hamilton Shirley Amerasinghe_sentence_6"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7040,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.09",
"value": "?search="
},
{
"operation": "search",
"time": "13.094",
"value": "Hamilton Shirley Amerasinghe"
},
{
"operation": "Now on",
"time": "14.182",
"value": "Hamilton Shirley Amerasinghe"
},
{
"operation": "Highlighting",
"time": "109.096",
"value": "Hamilton Shirley Amerasinghe_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "109.879",
"value": "Hamilton Shirley Amerasinghe_sentence_0"
},
{
"operation": "Highlighting",
"time": "117.563",
"value": "Hamilton Shirley Amerasinghe_header_cell_0_1_0"
},
{
"operation": "Highlighting deleted",
"time": "128.357",
"value": "Hamilton Shirley Amerasinghe_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "131.701",
"value": "Hamilton Shirley Amerasinghe_sentence_4"
},
{
"operation": "Highlighting",
"time": "135.779",
"value": "Hamilton Shirley Amerasinghe_sentence_5"
},
{
"operation": "Highlighting",
"time": "136.344",
"value": "Hamilton Shirley Amerasinghe_sentence_6"
},
{
"operation": "finish",
"time": "141.869",
"value": "finish"
}
]
}
|
exfever_6289
|
Lorelai Gilmore is a character in the WB dramedy television series Gilmore Girls,?portrayed by an American actress who received nominations for the Screen Actors Guild.
|
{
"evidence": [
"Lorelai Gilmore",
"Lauren Graham"
],
"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:\nLorelai Gilmore is a character in the WB dramedy television series Gilmore Girls,?portrayed by an American actress who received nominations for the Screen Actors Guild.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Lorelai Gilmore",
"Lauren Graham"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7041,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
74788
|
The Central Warwickshire Youth Football League had a representative side who did not win against the Stratford-upon-Avon League and Notts Elizabethen League; then in 1978, "Gordon Beard era" and the famous Olton Ravens started.
|
{
"evidence": [
"Central Warwickshire Youth Football League_sentence_24"
],
"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 Central Warwickshire Youth Football League had a representative side who did not win against the Stratford-upon-Avon League and Notts Elizabethen League; then in 1978, \"Gordon Beard era\" and the famous Olton Ravens started.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Central Warwickshire Youth Football League_sentence_24"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7042,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.957",
"value": "?search="
},
{
"operation": "search",
"time": "11.034",
"value": "Central Warwickshire Youth Football League"
},
{
"operation": "Now on",
"time": "13.697",
"value": "Central Warwickshire Youth Football League"
},
{
"operation": "Page search",
"time": "26.97",
"value": "representative"
},
{
"operation": "Highlighting",
"time": "43.386",
"value": "Central Warwickshire Youth Football League_sentence_24"
},
{
"operation": "finish",
"time": "59.839",
"value": "finish"
}
]
}
|
exfever_5185
|
The Battle of Corinth was fought between the Romans and the Greeks with allies from the Hellenistic era of the central peninsula and a geographic region in southern Greece.
|
{
"evidence": [
"Battle of Corinth (146 BC)",
"Achaean League",
"Peloponnese"
],
"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 Battle of Corinth was fought between the Romans and the Greeks with allies from the Hellenistic era of the central peninsula and a geographic region in southern Greece.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Battle of Corinth (146 BC)",
"Achaean League",
"Peloponnese"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7043,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
18044
|
Special Scout Connie Ryan (born on February 27, 1920) was named to finish out the string with only a handful of games left in the season; then for 1976, GM Eddie Robinson promised to hire a "firebrand" to replace the scholarly King.
|
{
"evidence": [
"Connie Ryan_sentence_0",
"Connie Ryan_sentence_6",
"Clyde King_sentence_25"
],
"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:\nSpecial Scout Connie Ryan (born on February 27, 1920) was named to finish out the string with only a handful of games left in the season; then for 1976, GM Eddie Robinson promised to hire a \"firebrand\" to replace the scholarly King.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Connie Ryan_sentence_0",
"Connie Ryan_sentence_6",
"Clyde King_sentence_25"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7044,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.668",
"value": "?search="
},
{
"operation": "search",
"time": "14.719",
"value": "Connie Ryan"
},
{
"operation": "Now on",
"time": "15.66",
"value": "Connie Ryan"
},
{
"operation": "Highlighting",
"time": "44.272",
"value": "Connie Ryan_sentence_0"
},
{
"operation": "Page search",
"time": "84.279",
"value": "string"
},
{
"operation": "Highlighting",
"time": "121.08",
"value": "Connie Ryan_sentence_6"
},
{
"operation": "Page search",
"time": "188.231",
"value": "eddie robinson"
},
{
"operation": "search",
"time": "206.161",
"value": "Eddie Robinson (baseball)"
},
{
"operation": "Now on",
"time": "207.366",
"value": "Eddie Robinson (baseball)"
},
{
"operation": "back-button-clicked",
"time": "265.673",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "265.78",
"value": "Connie Ryan"
},
{
"operation": "hyperlink",
"time": "270.871",
"value": "Clyde King"
},
{
"operation": "Now on",
"time": "271.669",
"value": "Clyde King"
},
{
"operation": "Highlighting",
"time": "297.454",
"value": "Clyde King_sentence_25"
},
{
"operation": "finish",
"time": "322.726",
"value": "finish"
}
]
}
|
85519
|
Manufactured by Stadler Rail, Variobahn, a German-designed model of articulated low-floor tram and light rail vehicle, comprises 25.71% of the total number of trams operating in Tramz.
|
{
"evidence": [
"Stadler Variobahn_sentence_0",
"Stadler Variobahn_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:\nManufactured by Stadler Rail, Variobahn, a German-designed model of articulated low-floor tram and light rail vehicle, comprises 25.71% of the total number of trams operating in Tramz.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Stadler Variobahn_sentence_0",
"Stadler Variobahn_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7045,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.082",
"value": "?search="
},
{
"operation": "search",
"time": "13.254",
"value": "Variobahn for TfL"
},
{
"operation": "Now on",
"time": "14.237",
"value": "Stadler Variobahn"
},
{
"operation": "Highlighting",
"time": "21.222",
"value": "Stadler Variobahn_sentence_0"
},
{
"operation": "Highlighting",
"time": "37.043",
"value": "Stadler Variobahn_sentence_1"
},
{
"operation": "Page search",
"time": "49.559",
"value": "tramz"
},
{
"operation": "Page search",
"time": "66.391",
"value": "25.71"
},
{
"operation": "Page search",
"time": "138.271",
"value": "25"
},
{
"operation": "finish",
"time": "151.457",
"value": "finish"
}
]
}
|
exfever_1296
|
Klang Valley is the area in central Selangor, Malaysia comprising Kuala Lumpur plus Putrajaya, and its surroundings and suburbs are naturally delineated by hilly areas and the Port Klang coastline.
|
{
"evidence": [
"Kuala Lumpur",
"Urban agglomeration"
],
"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:\nKlang Valley is the area in central Selangor, Malaysia comprising Kuala Lumpur plus Putrajaya, and its surroundings and suburbs are naturally delineated by hilly areas and the Port Klang coastline.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Kuala Lumpur",
"Urban agglomeration"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7046,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
52655
|
Grant Lightbown retired in 1991.
|
{
"evidence": [
"Grant Lightbown_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:\nGrant Lightbown retired in 1991.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Grant Lightbown_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7047,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.307",
"value": "?search="
},
{
"operation": "search",
"time": "24.631",
"value": "Grant Lightbown"
},
{
"operation": "Now on",
"time": "25.624",
"value": "Grant Lightbown"
},
{
"operation": "Highlighting",
"time": "287.9",
"value": "Grant Lightbown_sentence_1"
},
{
"operation": "finish",
"time": "355.154",
"value": "finish"
}
]
}
|
71814
|
Introduced in 1970, Triumph 1500, a small front-wheel drive car that was produced by Standard-Triumph from 1970 to 1973, is the first car engine of any brand to have a capacity of 1493 cc.
|
{
"evidence": [
"Triumph 1500_sentence_0",
"Triumph 1500_sentence_11",
"Triumph 1500_sentence_12"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nIntroduced in 1970, Triumph 1500, a small front-wheel drive car that was produced by Standard-Triumph from 1970 to 1973, is the first car engine of any brand to have a capacity of 1493 cc.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Triumph 1500_sentence_0",
"Triumph 1500_sentence_11",
"Triumph 1500_sentence_12"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7048,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.359",
"value": "?search="
},
{
"operation": "search",
"time": "9.989",
"value": "Triumph 1500"
},
{
"operation": "Now on",
"time": "11.199",
"value": "Triumph 1500"
},
{
"operation": "Highlighting",
"time": "14.373",
"value": "Triumph 1500_sentence_0"
},
{
"operation": "Highlighting",
"time": "60.587",
"value": "Triumph 1500_sentence_11"
},
{
"operation": "Highlighting",
"time": "61.782",
"value": "Triumph 1500_sentence_12"
},
{
"operation": "finish",
"time": "94.557",
"value": "finish"
}
]
}
|
70859
|
It was first opened to the public in the 1920s as a part of Clear Water Bay Road, yet it was split from the road and named after the Choi Hung Estate in 1963; and nowadays, the road has four lanes for traffic in both ways.
|
{
"evidence": [
"Choi Hung Road_sentence_1",
"Choi Hung Road_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nIt was first opened to the public in the 1920s as a part of Clear Water Bay Road, yet it was split from the road and named after the Choi Hung Estate in 1963; and nowadays, the road has four lanes for traffic in both ways.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Choi Hung Road_sentence_1",
"Choi Hung Road_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7049,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.121",
"value": "?search="
},
{
"operation": "search",
"time": "23.08",
"value": "Clear Water Bay Road,"
},
{
"operation": "Now on",
"time": "25.474",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "41.588",
"value": "Choi Hung Estate"
},
{
"operation": "Now on",
"time": "43.767",
"value": "Choi Hung Estate"
},
{
"operation": "Page search",
"time": "61.616",
"value": "clear water"
},
{
"operation": "search",
"time": "117.386",
"value": "choi hung road"
},
{
"operation": "Now on",
"time": "119.558",
"value": "Choi Hung Road"
},
{
"operation": "Highlighting",
"time": "122.732",
"value": "Choi Hung Road_sentence_1"
},
{
"operation": "Highlighting",
"time": "131.828",
"value": "Choi Hung Road_sentence_2"
},
{
"operation": "finish",
"time": "142.123",
"value": "finish"
}
]
}
|
exfever_6237
|
Joseph Vissarionovich Stalin was the longest-serving Soviet leader.
|
{
"evidence": [
"Joseph Stalin",
"General Secretary of the Communist Party of the Soviet Union"
],
"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:\nJoseph Vissarionovich Stalin was the longest-serving Soviet leader.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Joseph Stalin",
"General Secretary of the Communist Party of the Soviet Union"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7050,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
49599
|
Radford Semele is a village and civil parish in Warwickshire, England with a population of 2,012 as of 2011.
|
{
"evidence": [
"Radford Semele_sentence_0",
"Radford Semele_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:\nRadford Semele is a village and civil parish in Warwickshire, England with a population of 2,012 as of 2011.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Radford Semele_sentence_0",
"Radford Semele_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7051,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.852",
"value": "?search="
},
{
"operation": "search",
"time": "6.349",
"value": "Radford Semele"
},
{
"operation": "Now on",
"time": "7.231",
"value": "Radford Semele"
},
{
"operation": "Highlighting",
"time": "31.977",
"value": "Radford Semele_sentence_0"
},
{
"operation": "Highlighting",
"time": "40.993",
"value": "Radford Semele_sentence_1"
},
{
"operation": "finish",
"time": "269.977",
"value": "finish"
}
]
}
|
72803
|
United States' Vic Ghezzi and Byron Nelson faced each other in 18 holes at the 1941 PGA Championship.
|
{
"evidence": [
"1941 PGA Championship_sentence_1"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nUnited States' Vic Ghezzi and Byron Nelson faced each other in 18 holes at the 1941 PGA Championship.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"1941 PGA Championship_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7052,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.675",
"value": "?search="
},
{
"operation": "search",
"time": "10.441",
"value": "1941 PGA Championship"
},
{
"operation": "Now on",
"time": "11.832",
"value": "1941 PGA Championship"
},
{
"operation": "Page search",
"time": "28.51",
"value": "Vic Ghezzi"
},
{
"operation": "Highlighting",
"time": "33.841",
"value": "1941 PGA Championship_sentence_1"
},
{
"operation": "hyperlink",
"time": "61.248",
"value": "Vic Ghezzi"
},
{
"operation": "Now on",
"time": "62.033",
"value": "Vic Ghezzi"
},
{
"operation": "Highlighting",
"time": "67.872",
"value": "Vic Ghezzi_cell_0_7_1"
},
{
"operation": "Highlighting",
"time": "67.873",
"value": "Vic Ghezzi_title"
},
{
"operation": "Highlighting",
"time": "67.89",
"value": "Vic Ghezzi_header_cell_0_7_0"
},
{
"operation": "Highlighting",
"time": "67.89",
"value": "Vic Ghezzi_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "67.891",
"value": "Vic Ghezzi_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "77.503",
"value": "Vic Ghezzi_cell_0_7_1"
},
{
"operation": "Highlighting deleted",
"time": "77.505",
"value": "Vic Ghezzi_title"
},
{
"operation": "Highlighting deleted",
"time": "77.52",
"value": "Vic Ghezzi_header_cell_0_7_0"
},
{
"operation": "Highlighting deleted",
"time": "77.522",
"value": "Vic Ghezzi_header_cell_0_1_0"
},
{
"operation": "Highlighting deleted",
"time": "77.523",
"value": "Vic Ghezzi_header_cell_0_0_0"
},
{
"operation": "finish",
"time": "82.774",
"value": "finish"
}
]
}
|
36790
|
Initially, the Federation was located in Norsborg and had less than 3,000 members.
|
{
"evidence": [
"Syriacs in Sweden_sentence_29"
],
"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:\nInitially, the Federation was located in Norsborg and had less than 3,000 members.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Syriacs in Sweden_sentence_29"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7053,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.703",
"value": "?search="
},
{
"operation": "search",
"time": "28.335",
"value": "federation norsborg"
},
{
"operation": "Now on",
"time": "29.382",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "38.678",
"value": "Syriacs in Sweden"
},
{
"operation": "Now on",
"time": "39.3",
"value": "Syriacs in Sweden"
},
{
"operation": "Page search",
"time": "51.582",
"value": "federation"
},
{
"operation": "Highlighting",
"time": "75.307",
"value": "Syriacs in Sweden_sentence_26"
},
{
"operation": "Highlighting",
"time": "85.824",
"value": "Syriacs in Sweden_sentence_29"
},
{
"operation": "Highlighting deleted",
"time": "90.45",
"value": "Syriacs in Sweden_sentence_26"
},
{
"operation": "Highlighting",
"time": "112.846",
"value": "Syriacs in Sweden_sentence_30"
},
{
"operation": "Highlighting deleted",
"time": "146.315",
"value": "Syriacs in Sweden_sentence_30"
},
{
"operation": "finish",
"time": "149.76",
"value": "finish"
}
]
}
|
76358
|
In the Post-war era, 1950s and continuing through the 1980s, mechanization of the baking process required drier dough than hand-formed loaves, leading to drier loaves that do not have the same large gas bubbles and chewy consistency of good sourdough, and despite this trend, sourdough remained popular in the bay area.
|
{
"evidence": [
"History of California bread_sentence_20",
"History of California bread_sentence_18",
"History of California bread_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:\nIn the Post-war era, 1950s and continuing through the 1980s, mechanization of the baking process required drier dough than hand-formed loaves, leading to drier loaves that do not have the same large gas bubbles and chewy consistency of good sourdough, and despite this trend, sourdough remained popular in the bay area.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"History of California bread_sentence_20",
"History of California bread_sentence_18",
"History of California bread_sentence_21"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7054,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.959",
"value": "?search="
},
{
"operation": "search",
"time": "53.247",
"value": "mechanization of the baking process"
},
{
"operation": "Now on",
"time": "54.599",
"value": "undefined"
},
{
"operation": "search",
"time": "116.181",
"value": "History of California bread"
},
{
"operation": "Now on",
"time": "117.125",
"value": "History of California bread"
},
{
"operation": "Highlighting",
"time": "133.624",
"value": "History of California bread_sentence_20"
},
{
"operation": "Highlighting",
"time": "138.549",
"value": "History of California bread_sentence_18"
},
{
"operation": "Highlighting",
"time": "151.783",
"value": "History of California bread_sentence_21"
},
{
"operation": "finish",
"time": "156.159",
"value": "finish"
}
]
}
|
64471
|
According to Aleister Crowley, the last three Aeons within the system of Thelema have been (1) the Aeon of Isis, (2) the Aeon of Osiris, and (3) the current Aeon of Horus which began in 1904 with the writing of The Book of the Law
|
{
"evidence": [
"Aeon (Thelema)_sentence_31",
"Aeon (Thelema)_sentence_32",
"Aeon (Thelema)_sentence_7",
"Aeon (Thelema)_sentence_14",
"Aeon (Thelema)_sentence_18"
],
"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:\nAccording to Aleister Crowley, the last three Aeons within the system of Thelema have been (1) the Aeon of Isis, (2) the Aeon of Osiris, and (3) the current Aeon of Horus which began in 1904 with the writing of The Book of the Law\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Aeon (Thelema)_sentence_31",
"Aeon (Thelema)_sentence_32",
"Aeon (Thelema)_sentence_7",
"Aeon (Thelema)_sentence_14",
"Aeon (Thelema)_sentence_18"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7055,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.698",
"value": "?search="
},
{
"operation": "search",
"time": "23.733",
"value": "Contains...system of Thelema"
},
{
"operation": "Now on",
"time": "25.579",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "37.177",
"value": "Choronzon"
},
{
"operation": "Now on",
"time": "38.768",
"value": "Choronzon"
},
{
"operation": "search",
"time": "61.009",
"value": "Aleister Crowley"
},
{
"operation": "Now on",
"time": "78.509",
"value": "Aleister Crowley"
},
{
"operation": "Page search",
"time": "103.499",
"value": "aeon"
},
{
"operation": "hyperlink",
"time": "122.906",
"value": "Thelema"
},
{
"operation": "Now on",
"time": "132.65",
"value": "Thelema"
},
{
"operation": "search",
"time": "179.513",
"value": "Aeon of Osiris"
},
{
"operation": "Now on",
"time": "184.459",
"value": "Aeon (Thelema)"
},
{
"operation": "hyperlink",
"time": "255.935",
"value": "2 Differing views Aeon (Thelema)_section_5"
},
{
"operation": "Highlighting",
"time": "259.346",
"value": "Aeon (Thelema)_sentence_31"
},
{
"operation": "Highlighting",
"time": "260.551",
"value": "Aeon (Thelema)_sentence_32"
},
{
"operation": "search",
"time": "286.511",
"value": "crowley aeons"
},
{
"operation": "Now on",
"time": "288.614",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "303.806",
"value": "Aleister Crowley Thelema_section_9"
},
{
"operation": "Now on",
"time": "304.38",
"value": "Thelema#Aleister Crowley Thelema section 9"
},
{
"operation": "back-button-clicked",
"time": "344.838",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "345.504",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "355.826",
"value": "Aeonic Cosmology and Nazism Order of Nine Angles_section_5"
},
{
"operation": "Now on",
"time": "365.91",
"value": "Order of Nine Angles#Aeonic Cosmology and Nazism Order of Nine Angles section 5"
},
{
"operation": "Page search",
"time": "373.918",
"value": "crowley"
},
{
"operation": "hyperlink",
"time": "401.051",
"value": "Thelemic"
},
{
"operation": "Now on",
"time": "401.637",
"value": "Thelema"
},
{
"operation": "Page search",
"time": "410.381",
"value": "aeon*"
},
{
"operation": "search",
"time": "468.439",
"value": "Aeon of Horus"
},
{
"operation": "Now on",
"time": "469.105",
"value": "Aeon (Thelema)"
},
{
"operation": "Highlighting",
"time": "490.643",
"value": "Aeon (Thelema)_sentence_7"
},
{
"operation": "Highlighting",
"time": "495.57",
"value": "Aeon (Thelema)_sentence_14"
},
{
"operation": "Highlighting",
"time": "499.299",
"value": "Aeon (Thelema)_sentence_18"
},
{
"operation": "finish",
"time": "524.941",
"value": "finish"
}
]
}
|
exfever_1887
|
Man of Steel is a 2013 superhero film directed by an American film director and producer who made his feature film debut in Dawn of the Dead.
|
{
"evidence": [
"Man of Steel (film)",
"Zack Snyder"
],
"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:\nMan of Steel is a 2013 superhero film directed by an American film director and producer who made his feature film debut in Dawn of the Dead.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Man of Steel (film)",
"Zack Snyder"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7056,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
43310
|
The National Nordic Museum is a heritage museum in UK that was established in 1980.
|
{
"evidence": [
"National Nordic Museum_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 National Nordic Museum is a heritage museum in UK that was established in 1980.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"National Nordic Museum_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7057,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.607",
"value": "?search="
},
{
"operation": "search",
"time": "12.102",
"value": "National Nordic Museum"
},
{
"operation": "Now on",
"time": "12.96",
"value": "National Nordic Museum"
},
{
"operation": "Highlighting",
"time": "24.333",
"value": "National Nordic Museum_cell_0_3_1"
},
{
"operation": "Highlighting",
"time": "24.335",
"value": "National Nordic Museum_header_cell_0_3_0"
},
{
"operation": "Highlighting",
"time": "27.761",
"value": "National Nordic Museum_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "37.688",
"value": "National Nordic Museum_cell_0_3_1"
},
{
"operation": "Highlighting deleted",
"time": "37.691",
"value": "National Nordic Museum_header_cell_0_3_0"
},
{
"operation": "finish",
"time": "57.176",
"value": "finish"
}
]
}
|
exfever_4808
|
San Junipero’s fourth episode had higher critical ratings than the other episodes.
|
{
"evidence": [
"San Junipero",
"Charlie Brooker"
],
"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:\nSan Junipero’s fourth episode had higher critical ratings than the other episodes.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"San Junipero",
"Charlie Brooker"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7058,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3652
|
Pulp Fiction is a comedy crime film starring an actor who appeared in a science fiction action film directed by Steven Spielberg.
|
{
"evidence": [
"Pulp Fiction",
"Samuel L. Jackson",
"Jurassic Park (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:\nPulp Fiction is a comedy crime film starring an actor who appeared in a science fiction action film directed by Steven Spielberg.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Pulp Fiction",
"Samuel L. Jackson",
"Jurassic Park (film)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7059,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_497
|
James Blake Litherland is an English multi-instrumentalist and record producer who collaborated with an American indie-folk band founded in 2009 by singer-songwriter Justin Vernon.
|
{
"evidence": [
"James Blake (musician)",
"Bon Iver"
],
"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:\nJames Blake Litherland is an English multi-instrumentalist and record producer who collaborated with an American indie-folk band founded in 2009 by singer-songwriter Justin Vernon.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"James Blake (musician)",
"Bon Iver"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7060,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5799
|
Rajon Pierre Rondo is an American professional baseball player who was drafted 21st overall by the Phoenix Suns in the NBA draft that was held on June 28, 2006, at the Theatre at Madison Square Garden in New York City.
|
{
"evidence": [
"Rajon Rondo",
"2006 NBA draft"
],
"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:\nRajon Pierre Rondo is an American professional baseball player who was drafted 21st overall by the Phoenix Suns in the NBA draft that was held on June 28, 2006, at the Theatre at Madison Square Garden in New York City.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Rajon Rondo",
"2006 NBA draft"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7061,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
51758
|
Ornithogalum lebaense is often found growing with Aloe mendesii, a plant that is also an obligatory cremnophyte of the area and O. lebaense flowers in the winter and is also winter deciduous "falling off at maturity" and "tending to fall off".
|
{
"evidence": [
"Ornithogalum lebaense_sentence_9",
"Ornithogalum lebaense_sentence_10",
"Deciduous_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:\nOrnithogalum lebaense is often found growing with Aloe mendesii, a plant that is also an obligatory cremnophyte of the area and O. lebaense flowers in the winter and is also winter deciduous \"falling off at maturity\" and \"tending to fall off\".\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Ornithogalum lebaense_sentence_9",
"Ornithogalum lebaense_sentence_10",
"Deciduous_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7062,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.652",
"value": "?search="
},
{
"operation": "search",
"time": "12.502",
"value": "Ornithogalum lebaense"
},
{
"operation": "Now on",
"time": "13.4",
"value": "Ornithogalum lebaense"
},
{
"operation": "Highlighting",
"time": "46.779",
"value": "Ornithogalum lebaense_sentence_9"
},
{
"operation": "Highlighting",
"time": "55.672",
"value": "Ornithogalum lebaense_sentence_10"
},
{
"operation": "hyperlink",
"time": "76.77",
"value": "deciduous"
},
{
"operation": "Now on",
"time": "77.566",
"value": "Deciduous"
},
{
"operation": "Highlighting",
"time": "80.694",
"value": "Deciduous_sentence_1"
},
{
"operation": "finish",
"time": "92.022",
"value": "finish"
}
]
}
|
exfever_134
|
Tommy Lee Jones' fifth film is Men in Black.
|
{
"evidence": [
"Tommy Lee Jones",
"The Fugitive (1993 film)"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nTommy Lee Jones' fifth film is Men in Black.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tommy Lee Jones",
"The Fugitive (1993 film)"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7063,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
3668
|
The 1919 Lineup for the Antonio Black Aces included Right Fielder Grant Don and Shortstop Namon Washington, who only played with the Antonio Black Aces during his career.
|
{
"evidence": [
"Namon Washington_sentence_1"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe 1919 Lineup for the Antonio Black Aces included Right Fielder Grant Don and Shortstop Namon Washington, who only played with the Antonio Black Aces during his career.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Namon Washington_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7064,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "search",
"time": "13.815",
"value": "1919 Antonio Black Aces"
},
{
"operation": "hyperlink",
"time": "18.923",
"value": "San Antonio Black Aces"
},
{
"operation": "Highlighting",
"time": "36.056",
"value": "San Antonio Black Aces_cell_1_11_1"
},
{
"operation": "Highlighting",
"time": "36.058",
"value": "San Antonio Black Aces_title"
},
{
"operation": "Highlighting",
"time": "36.06",
"value": "San Antonio Black Aces_section_0"
},
{
"operation": "Highlighting",
"time": "36.064",
"value": "San Antonio Black Aces_header_cell_1_0_1"
},
{
"operation": "Highlighting",
"time": "38.101",
"value": "San Antonio Black Aces_cell_1_11_0"
},
{
"operation": "Highlighting",
"time": "38.113",
"value": "San Antonio Black Aces_header_cell_1_0_0"
},
{
"operation": "Highlighting",
"time": "39.429",
"value": "San Antonio Black Aces_cell_1_9_1"
},
{
"operation": "Highlighting",
"time": "40.903",
"value": "San Antonio Black Aces_cell_1_9_0"
},
{
"operation": "hyperlink",
"time": "47.642",
"value": "Namon Washington"
},
{
"operation": "Highlighting",
"time": "51.82",
"value": "Namon Washington_sentence_1"
},
{
"operation": "finish",
"time": "86.715",
"value": "finish"
}
]
}
|
exfever_1509
|
My Own Private Idaho is a 1991 American feature film that was produced outside of the major film studio system.
|
{
"evidence": [
"My Own Private Idaho",
"Independent 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:\nMy Own Private Idaho is a 1991 American feature film that was produced outside of the major film studio system.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"My Own Private Idaho",
"Independent film"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7065,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
93851
|
Israel Lund is a conceptual painter based in Brooklyn, New York who creates acrylic paintings using a combination of digital and analog techniques.
|
{
"evidence": [
"Israel Lund_sentence_0",
"Israel Lund_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:\nIsrael Lund is a conceptual painter based in Brooklyn, New York who creates acrylic paintings using a combination of digital and analog techniques.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Israel Lund_sentence_0",
"Israel Lund_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7066,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.282",
"value": "?search="
},
{
"operation": "search",
"time": "12.713",
"value": "Israel Lund"
},
{
"operation": "Now on",
"time": "13.379",
"value": "Israel Lund"
},
{
"operation": "Highlighting",
"time": "16.211",
"value": "Israel Lund_sentence_0"
},
{
"operation": "Highlighting",
"time": "25.025",
"value": "Israel Lund_sentence_1"
},
{
"operation": "finish",
"time": "33.02",
"value": "finish"
}
]
}
|
26296
|
The College Dropout Video Anthology first video was rejected by MTV and a third video was made but is rarely seen.
|
{
"evidence": [
"The College Dropout Video Anthology_sentence_4",
"The College Dropout Video Anthology_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:\nThe College Dropout Video Anthology first video was rejected by MTV and a third video was made but is rarely seen.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"The College Dropout Video Anthology_sentence_4",
"The College Dropout Video Anthology_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7067,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.131",
"value": "?search="
},
{
"operation": "search",
"time": "40.922",
"value": "College Dropout Video Anthology"
},
{
"operation": "Now on",
"time": "42.353",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "46.66",
"value": "The College Dropout Video Anthology"
},
{
"operation": "Now on",
"time": "47.207",
"value": "The College Dropout Video Anthology"
},
{
"operation": "Page search",
"time": "72.21",
"value": "mtv"
},
{
"operation": "Highlighting",
"time": "79.885",
"value": "The College Dropout Video Anthology_sentence_4"
},
{
"operation": "Highlighting",
"time": "84.314",
"value": "The College Dropout Video Anthology_sentence_5"
},
{
"operation": "finish",
"time": "93.22",
"value": "finish"
}
]
}
|
exfever_6514
|
"Live Your Life" is a song by American rapper T.I. which both samples and interpolates the 2004 song "Dragostea Din Tei" by a group that originated in 1998 as a duo consisting of Petru Jelihovschi and a Moldovan artist, songwriter, and record producer.
|
{
"evidence": [
"Live Your Life (T.I. song)",
"O-Zone",
"Dan Balan"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\n\"Live Your Life\" is a song by American rapper T.I. which both samples and interpolates the 2004 song \"Dragostea Din Tei\" by a group that originated in 1998 as a duo consisting of Petru Jelihovschi and a Moldovan artist, songwriter, and record producer.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Live Your Life (T.I. song)",
"O-Zone",
"Dan Balan"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7068,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
40682
|
Cheryl McKay is an American author, screenwriter and winner of the 2006/2007 Art Within Labs award from California who wrote the screen adaptations of Jim Stovall's novel The Ultimate Gift and Sheila Walsh's children's book Gigi: God's Little Princess and Taylor's Wall and also wrote The Wild & Wacky Totally True Bible Stories series, narrated by Frank Peretti and the ACFW/Carol Award winning novel, Never the Bride.
|
{
"evidence": [
"Cheryl McKay_sentence_0",
"Cheryl McKay_sentence_1",
"Cheryl McKay_sentence_2",
"Cheryl McKay_sentence_3",
"Cheryl McKay_sentence_4"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nCheryl McKay is an American author, screenwriter and winner of the 2006/2007 Art Within Labs award from California who wrote the screen adaptations of Jim Stovall's novel The Ultimate Gift and Sheila Walsh's children's book Gigi: God's Little Princess and Taylor's Wall and also wrote The Wild & Wacky Totally True Bible Stories series, narrated by Frank Peretti and the ACFW/Carol Award winning novel, Never the Bride.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Cheryl McKay_sentence_0",
"Cheryl McKay_sentence_1",
"Cheryl McKay_sentence_2",
"Cheryl McKay_sentence_3",
"Cheryl McKay_sentence_4"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7069,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.84",
"value": "undefined"
},
{
"operation": "search",
"time": "36.245",
"value": "Cheryl McKay"
},
{
"operation": "Now on",
"time": "37.388",
"value": "Cheryl McKay"
},
{
"operation": "Highlighting",
"time": "50.785",
"value": "Cheryl McKay_sentence_0"
},
{
"operation": "Highlighting",
"time": "51.698",
"value": "Cheryl McKay_sentence_1"
},
{
"operation": "Highlighting",
"time": "106.253",
"value": "Cheryl McKay_sentence_2"
},
{
"operation": "Highlighting",
"time": "107.543",
"value": "Cheryl McKay_sentence_3"
},
{
"operation": "Highlighting",
"time": "167.906",
"value": "Cheryl McKay_sentence_4"
},
{
"operation": "search",
"time": "187.985",
"value": "Never the Bride"
},
{
"operation": "Now on",
"time": "188.971",
"value": "Never the Bride"
},
{
"operation": "search",
"time": "213.127",
"value": "Never the Bride novel"
},
{
"operation": "Now on",
"time": "214.861",
"value": "undefined"
},
{
"operation": "search",
"time": "233.389",
"value": "ACFW/Carol Award winning novel, Never the Bride"
},
{
"operation": "Now on",
"time": "234.281",
"value": "undefined"
},
{
"operation": "search",
"time": "243.743",
"value": "ACFW/Carol Award"
},
{
"operation": "Now on",
"time": "244.377",
"value": "undefined"
},
{
"operation": "finish",
"time": "316.737",
"value": "finish"
}
]
}
|
18077
|
The literacy rate of Mithi Chhapri in the Bhopal District of Madhya Pradesh, India is 35% among the women population and 59% among men, based on the 2011 Census.
|
{
"evidence": [
"Mithi Chhapri_sentence_0",
"Mithi Chhapri_sentence_3"
],
"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 literacy rate of Mithi Chhapri in the Bhopal District of Madhya Pradesh, India is 35% among the women population and 59% among men, based on the 2011 Census.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Mithi Chhapri_sentence_0",
"Mithi Chhapri_sentence_3"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7070,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.538",
"value": "?search="
},
{
"operation": "search",
"time": "22.636",
"value": "The literacy rate of Mithi Chhapri"
},
{
"operation": "Now on",
"time": "25.367",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "27.284",
"value": "Mithi Chhapri"
},
{
"operation": "Now on",
"time": "28.586",
"value": "Mithi Chhapri"
},
{
"operation": "Highlighting",
"time": "44.188",
"value": "Mithi Chhapri_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "52.09",
"value": "Mithi Chhapri_sentence_0"
},
{
"operation": "Highlighting",
"time": "53.905",
"value": "Mithi Chhapri_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "65.103",
"value": "Mithi Chhapri_sentence_3"
},
{
"operation": "Highlighting",
"time": "123.05",
"value": "Mithi Chhapri_sentence_1"
},
{
"operation": "Highlighting",
"time": "123.631",
"value": "Mithi Chhapri_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "124.404",
"value": "Mithi Chhapri_sentence_1"
},
{
"operation": "hyperlink",
"time": "127.845",
"value": "2011 census of India"
},
{
"operation": "Now on",
"time": "131.858",
"value": "2011 Census of India"
},
{
"operation": "Page search",
"time": "143.339",
"value": "Mithi Chhapri"
},
{
"operation": "hyperlink",
"time": "192.542",
"value": "Madhya Pradesh"
},
{
"operation": "Now on",
"time": "196.661",
"value": "Madhya Pradesh"
},
{
"operation": "Page search",
"time": "202.907",
"value": "Mithi Chhapri"
},
{
"operation": "search",
"time": "228.134",
"value": "Mithi Chhapri"
},
{
"operation": "Now on",
"time": "229.728",
"value": "Mithi Chhapri"
},
{
"operation": "Highlighting",
"time": "239.84",
"value": "Mithi Chhapri_sentence_3"
},
{
"operation": "finish",
"time": "300.265",
"value": "finish"
}
]
}
|
94314
|
The very first German registration plates, from 1913 to the end of the 1920s, were rectangular, with a white background and with the name or initials of the colony in red followed by the registration number; however, special military service plates were used with the initials SOM (Somalia) or T (Tripolitania) in front.
|
{
"evidence": [
"Vehicle registration plates of the German Democratic Republic_sentence_0",
"Vehicle registration plates of the German Democratic Republic_sentence_1",
"Vehicle registration plates of the German Democratic Republic_sentence_2"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe very first German registration plates, from 1913 to the end of the 1920s, were rectangular, with a white background and with the name or initials of the colony in red followed by the registration number; however, special military service plates were used with the initials SOM (Somalia) or T (Tripolitania) in front.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Vehicle registration plates of the German Democratic Republic_sentence_0",
"Vehicle registration plates of the German Democratic Republic_sentence_1",
"Vehicle registration plates of the German Democratic Republic_sentence_2"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7071,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.538",
"value": "?search="
},
{
"operation": "Now on",
"time": "10.861",
"value": "?search="
},
{
"operation": "search",
"time": "16.857",
"value": "Contains...German registration plates"
},
{
"operation": "Now on",
"time": "18.628",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "21.252",
"value": "Vehicle registration plates of the German Democratic Republic"
},
{
"operation": "Now on",
"time": "22.431",
"value": "Vehicle registration plates of the German Democratic Republic"
},
{
"operation": "back-button-clicked",
"time": "31.319",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "32.298",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "65.563",
"value": "Special Plates Vehicle registration plates of the German Democratic Republic_section_2"
},
{
"operation": "Now on",
"time": "66.541",
"value": "Vehicle registration plates of the German Democratic Republic#Special Plates Vehicle registration plates of the German Democratic Republic section 2"
},
{
"operation": "hyperlink",
"time": "88.128",
"value": "German Democratic Republic"
},
{
"operation": "Now on",
"time": "92.78",
"value": "East Germany"
},
{
"operation": "back-button-clicked",
"time": "96.28",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "97.11",
"value": "Vehicle registration plates of the German Democratic Republic#Special Plates Vehicle registration plates of the German Democratic Republic section 2"
},
{
"operation": "hyperlink",
"time": "100.797",
"value": "1 Design Vehicle registration plates of the German Democratic Republic_section_0"
},
{
"operation": "Highlighting",
"time": "105.827",
"value": "Vehicle registration plates of the German Democratic Republic_sentence_0"
},
{
"operation": "Highlighting",
"time": "106.66",
"value": "Vehicle registration plates of the German Democratic Republic_sentence_1"
},
{
"operation": "Highlighting",
"time": "107.108",
"value": "Vehicle registration plates of the German Democratic Republic_sentence_2"
},
{
"operation": "finish",
"time": "113.211",
"value": "finish"
}
]
}
|
exfever_5164
|
The Godfather Part III is a 1990 American crime film starring talented artists, including is an American former actress born on January 27, 1964.
|
{
"evidence": [
"The Godfather Part III",
"Bridget Fonda"
],
"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 Godfather Part III is a 1990 American crime film starring talented artists, including is an American former actress born on January 27, 1964.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Godfather Part III",
"Bridget Fonda"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7072,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2799
|
The neocortex is also the newest part of the cerebral cortex to evolve.
|
{
"evidence": [
"Mammal",
"Neocortex"
],
"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 neocortex is also the newest part of the cerebral cortex to evolve.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Mammal",
"Neocortex"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7073,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
55014
|
East Kent Mercury is a weekly newspaper, owned by KM Group which serves Dover.
|
{
"evidence": [
"East Kent Mercury and Dover Mercury_sentence_0",
"East Kent Mercury and Dover Mercury_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nEast Kent Mercury is a weekly newspaper, owned by KM Group which serves Dover.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"East Kent Mercury and Dover Mercury_sentence_0",
"East Kent Mercury and Dover Mercury_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7074,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.652",
"value": "undefined"
},
{
"operation": "search",
"time": "11.291",
"value": "East Kent Mercury and Dover Mercury"
},
{
"operation": "Now on",
"time": "12.041",
"value": "East Kent Mercury and Dover Mercury"
},
{
"operation": "Highlighting",
"time": "142.201",
"value": "East Kent Mercury and Dover Mercury_sentence_0"
},
{
"operation": "Highlighting",
"time": "150.208",
"value": "East Kent Mercury and Dover Mercury_sentence_2"
},
{
"operation": "Highlighting",
"time": "155.178",
"value": "East Kent Mercury and Dover Mercury_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "158.64",
"value": "East Kent Mercury and Dover Mercury_sentence_1"
},
{
"operation": "hyperlink",
"time": "166.595",
"value": "KM Group"
},
{
"operation": "Now on",
"time": "168.188",
"value": "KM Group"
},
{
"operation": "back-button-clicked",
"time": "177.673",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "177.749",
"value": "East Kent Mercury and Dover Mercury"
},
{
"operation": "hyperlink",
"time": "187.999",
"value": "KM Group"
},
{
"operation": "Now on",
"time": "189.936",
"value": "KM Group"
},
{
"operation": "back-button-clicked",
"time": "210.294",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "210.354",
"value": "East Kent Mercury and Dover Mercury"
},
{
"operation": "hyperlink",
"time": "228.609",
"value": "KM Group"
},
{
"operation": "Now on",
"time": "229.363",
"value": "KM Group"
},
{
"operation": "back-button-clicked",
"time": "231.184",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "231.246",
"value": "East Kent Mercury and Dover Mercury"
},
{
"operation": "hyperlink",
"time": "248.178",
"value": "Deal"
},
{
"operation": "Now on",
"time": "249.104",
"value": "Deal, Kent"
},
{
"operation": "back-button-clicked",
"time": "255.81",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "255.875",
"value": "East Kent Mercury and Dover Mercury"
},
{
"operation": "hyperlink",
"time": "262.448",
"value": "KM Group"
},
{
"operation": "Now on",
"time": "262.769",
"value": "KM Group"
},
{
"operation": "back-button-clicked",
"time": "270.911",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "270.974",
"value": "East Kent Mercury and Dover Mercury"
},
{
"operation": "hyperlink",
"time": "321.129",
"value": "KM Group"
},
{
"operation": "Now on",
"time": "321.74",
"value": "KM Group"
},
{
"operation": "back-button-clicked",
"time": "335.499",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "335.557",
"value": "East Kent Mercury and Dover Mercury"
},
{
"operation": "finish",
"time": "338.723",
"value": "finish"
}
]
}
|
43121
|
Sarah Labowitz, co-founder and co-director of the NYU Stern Center for Business and Human Rights, earned her MA from New York University.
|
{
"evidence": [
"Sarah Labowitz_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:\nSarah Labowitz, co-founder and co-director of the NYU Stern Center for Business and Human Rights, earned her MA from New York University.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Sarah Labowitz_sentence_7"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7075,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.079",
"value": "?search="
},
{
"operation": "search",
"time": "7.692",
"value": "Sarah Labowitz"
},
{
"operation": "Now on",
"time": "8.892",
"value": "Sarah Labowitz"
},
{
"operation": "Highlighting",
"time": "19.091",
"value": "Sarah Labowitz_sentence_0"
},
{
"operation": "Highlighting",
"time": "30.275",
"value": "Sarah Labowitz_sentence_7"
},
{
"operation": "hyperlink",
"time": "32.341",
"value": "Tufts University"
},
{
"operation": "Now on",
"time": "34.001",
"value": "Tufts University"
},
{
"operation": "back-button-clicked",
"time": "42.877",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "42.999",
"value": "Sarah Labowitz"
},
{
"operation": "finish",
"time": "63.184",
"value": "finish"
}
]
}
|
exfever_2424
|
Gilmore Girls is an American comedy-drama television series that debuted on an American television network first launched on broadcast television on January 11, 1995.
|
{
"evidence": [
"Gilmore Girls",
"The WB"
],
"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:\nGilmore Girls is an American comedy-drama television series that debuted on an American television network first launched on broadcast television on January 11, 1995.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Gilmore Girls",
"The WB"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7076,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1456
|
The Adjustment Bureau is a 2011 American science fiction romantic thriller film that stars various actors, including an English actor.
|
{
"evidence": [
"The Adjustment Bureau",
"Terence Stamp"
],
"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 Adjustment Bureau is a 2011 American science fiction romantic thriller film that stars various actors, including an English actor.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Adjustment Bureau",
"Terence Stamp"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7077,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6642
|
Leonard Norman Cohen was a Canadian singer-songwriter, poet and novelist who was inducted into the Canadian Music Hall of Fame and in the Canadian non-profit organization founded in 1998 by Frank Davies.
|
{
"evidence": [
"Leonard Cohen",
"Canadian Songwriters Hall of Fame"
],
"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:\nLeonard Norman Cohen was a Canadian singer-songwriter, poet and novelist who was inducted into the Canadian Music Hall of Fame and in the Canadian non-profit organization founded in 1998 by Frank Davies.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Leonard Cohen",
"Canadian Songwriters Hall of Fame"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7078,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7143
|
France is a transcontinental country spanning Eastern Europe that uses the Romance language of the Indo-European family.
|
{
"evidence": [
"France",
"French language"
],
"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:\nFrance is a transcontinental country spanning Eastern Europe that uses the Romance language of the Indo-European family.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"France",
"French language"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7079,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
63922
|
The first game of the 1998 Philadelphia Phillies season, the franchise's 116th, was a 1-0 loss against the New York Mets.
|
{
"evidence": [
"1998 Philadelphia Phillies season_sentence_0",
"1998 Philadelphia Phillies season_sentence_1",
"1998 Philadelphia Phillies season_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:\nThe first game of the 1998 Philadelphia Phillies season, the franchise's 116th, was a 1-0 loss against the New York Mets.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"1998 Philadelphia Phillies season_sentence_0",
"1998 Philadelphia Phillies season_sentence_1",
"1998 Philadelphia Phillies season_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7080,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.74",
"value": "?search="
},
{
"operation": "search",
"time": "11.518",
"value": "1998 Philadelphia Phillies season"
},
{
"operation": "Now on",
"time": "12.841",
"value": "1998 Philadelphia Phillies season"
},
{
"operation": "Highlighting",
"time": "18.287",
"value": "1998 Philadelphia Phillies season_sentence_0"
},
{
"operation": "Highlighting",
"time": "45.044",
"value": "1998 Philadelphia Phillies season_sentence_1"
},
{
"operation": "Highlighting",
"time": "54.413",
"value": "1998 Philadelphia Phillies season_sentence_3"
},
{
"operation": "finish",
"time": "80.612",
"value": "finish"
}
]
}
|
25513
|
Diuris arenaria (Tomaree donkey orchid) is a tuberous herb with up to nine blue flowers, and the sepal is more or less erect.
|
{
"evidence": [
"Diuris arenaria_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:\nDiuris arenaria (Tomaree donkey orchid) is a tuberous herb with up to nine blue flowers, and the sepal is more or less erect.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Diuris arenaria_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7081,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.245",
"value": "?search="
},
{
"operation": "search",
"time": "11.132",
"value": "Diuris arenaria"
},
{
"operation": "Now on",
"time": "12.38",
"value": "Diuris arenaria"
},
{
"operation": "Highlighting",
"time": "19.045",
"value": "Diuris arenaria_sentence_0"
},
{
"operation": "hyperlink",
"time": "37.613",
"value": "orchid"
},
{
"operation": "Now on",
"time": "39.941",
"value": "Orchidaceae"
},
{
"operation": "Highlighting",
"time": "49.645",
"value": "Orchidaceae_sentence_2"
},
{
"operation": "Highlighting",
"time": "49.646",
"value": "Orchidaceae_title"
},
{
"operation": "back-button-clicked",
"time": "50.988",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "51.101",
"value": "Diuris arenaria"
},
{
"operation": "Highlighting",
"time": "58.622",
"value": "Diuris arenaria_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "60.159",
"value": "Diuris arenaria_sentence_0"
},
{
"operation": "Highlighting",
"time": "70.387",
"value": "Diuris arenaria_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "78.269",
"value": "Diuris arenaria_sentence_3"
},
{
"operation": "finish",
"time": "99.088",
"value": "finish"
}
]
}
|
exfever_4965
|
Diana Ross rose to fame as the lead singer of an American female singing group that rivaled an English rock band, formed in Liverpool in 1960.
|
{
"evidence": [
"Diana Ross",
"The Supremes",
"The Beatles"
],
"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:\nDiana Ross rose to fame as the lead singer of an American female singing group that rivaled an English rock band, formed in Liverpool in 1960.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Diana Ross",
"The Supremes",
"The Beatles"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 7082,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
71541
|
The SWAC commissioner announced that the game had been officially sanctioned, however the bowl was not revived and was the conference's of Division II playoffs.
|
{
"evidence": [
"SWAC Championship Game_sentence_16",
"SWAC Championship Game_sentence_17"
],
"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 SWAC commissioner announced that the game had been officially sanctioned, however the bowl was not revived and was the conference's of Division II playoffs.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"SWAC Championship Game_sentence_16",
"SWAC Championship Game_sentence_17"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 7083,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.168",
"value": "?search="
},
{
"operation": "search",
"time": "21.155",
"value": "Contains...SWAC commissioner"
},
{
"operation": "Now on",
"time": "22.517",
"value": "undefined"
},
{
"operation": "search",
"time": "30.502",
"value": "SWAC Championship Game"
},
{
"operation": "Now on",
"time": "32.632",
"value": "SWAC Championship Game"
},
{
"operation": "Highlighting",
"time": "192.591",
"value": "SWAC Championship Game_sentence_16"
},
{
"operation": "Highlighting",
"time": "198.127",
"value": "SWAC Championship Game_sentence_17"
},
{
"operation": "finish",
"time": "216.432",
"value": "finish"
}
]
}
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.