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_5215
|
Matthew Staton Bomer's last film is The Boys in the Band.
|
{
"evidence": [
"Matt Bomer",
"Primetime Emmy Award"
],
"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:\nMatthew Staton Bomer's last film is The Boys in the Band.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Matt Bomer",
"Primetime Emmy Award"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2684,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
23069
|
Cheese Lane Shot Tower was built in 1782 in Redcliffe, it was owned by the Sheldon Bush and Patent Shot Company Limited before it was demolished as part of a road widening scheme.
|
{
"evidence": [
"Cheese Lane Shot Tower_sentence_4",
"Cheese Lane Shot Tower_sentence_5",
"Cheese Lane Shot Tower_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:\nCheese Lane Shot Tower was built in 1782 in Redcliffe, it was owned by the Sheldon Bush and Patent Shot Company Limited before it was demolished as part of a road widening scheme.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Cheese Lane Shot Tower_sentence_4",
"Cheese Lane Shot Tower_sentence_5",
"Cheese Lane Shot Tower_sentence_6"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2685,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.251",
"value": "?search="
},
{
"operation": "search",
"time": "9.649",
"value": "Cheese Lane Shot Tower"
},
{
"operation": "Now on",
"time": "10.734",
"value": "Cheese Lane Shot Tower"
},
{
"operation": "Highlighting",
"time": "16.547",
"value": "Cheese Lane Shot Tower_sentence_4"
},
{
"operation": "Highlighting",
"time": "31.463",
"value": "Cheese Lane Shot Tower_sentence_5"
},
{
"operation": "Highlighting",
"time": "36.679",
"value": "Cheese Lane Shot Tower_sentence_6"
},
{
"operation": "finish",
"time": "42.889",
"value": "finish"
}
]
}
|
45961
|
The Engineering and Technology History Wiki (ETHW) is a MediaWiki-based website dedicated to the history of technology and information on how to submit an additional IEEE Milestones can be found on a special website and There is also a collection of "Landmarks" chosen by ASME as mechanical engineering highlights.
|
{
"evidence": [
"Engineering and Technology History Wiki_sentence_0",
"Engineering and Technology History Wiki_sentence_19",
"Engineering and Technology History Wiki_sentence_20"
],
"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 Engineering and Technology History Wiki (ETHW) is a MediaWiki-based website dedicated to the history of technology and information on how to submit an additional IEEE Milestones can be found on a special website and There is also a collection of \"Landmarks\" chosen by ASME as mechanical engineering highlights.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Engineering and Technology History Wiki_sentence_0",
"Engineering and Technology History Wiki_sentence_19",
"Engineering and Technology History Wiki_sentence_20"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2686,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.734",
"value": "?search="
},
{
"operation": "search",
"time": "12.483",
"value": "Engineering and Technology History Wiki"
},
{
"operation": "Now on",
"time": "13.31",
"value": "Engineering and Technology History Wiki"
},
{
"operation": "Highlighting",
"time": "19.468",
"value": "Engineering and Technology History Wiki_sentence_0"
},
{
"operation": "Page search",
"time": "41.232",
"value": "submit"
},
{
"operation": "Highlighting",
"time": "52.496",
"value": "Engineering and Technology History Wiki_sentence_19"
},
{
"operation": "search",
"time": "77.397",
"value": "ieee"
},
{
"operation": "Now on",
"time": "78.49",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "84.89",
"value": "Federal Information Processing Standards"
},
{
"operation": "Now on",
"time": "85.391",
"value": "Federal Information Processing Standards"
},
{
"operation": "back-button-clicked",
"time": "89.701",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "90.434",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "94.12",
"value": "IEEE 802.15.1"
},
{
"operation": "Now on",
"time": "95.815",
"value": "Bluetooth"
},
{
"operation": "search",
"time": "114.731",
"value": "ieee"
},
{
"operation": "Now on",
"time": "115.447",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "120.088",
"value": "Professional"
},
{
"operation": "Now on",
"time": "120.581",
"value": "Professional"
},
{
"operation": "Page search",
"time": "131.522",
"value": "ieee"
},
{
"operation": "back-button-clicked",
"time": "140.113",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "140.883",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "145.763",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "145.776",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "147.476",
"value": "Bluetooth"
},
{
"operation": "back-button-clicked",
"time": "151.858",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "151.872",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "152.628",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "153.869",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "153.882",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "153.896",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "154.21",
"value": "Engineering and Technology History Wiki"
},
{
"operation": "Page search",
"time": "188.55",
"value": "landmark"
},
{
"operation": "Highlighting",
"time": "203.883",
"value": "Engineering and Technology History Wiki_sentence_20"
},
{
"operation": "finish",
"time": "410.522",
"value": "finish"
}
]
}
|
exfever_4332
|
Christopher McQuarrie is a stage director.
|
{
"evidence": [
"Edge of Tomorrow",
"Action 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:\nChristopher McQuarrie is a stage director.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Edge of Tomorrow",
"Action film"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2687,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4973
|
The Golden Girls is an American movie about four older women who share a home in a state located in the Southeastern region bordered by the 24th-largest area in the U.S.
|
{
"evidence": [
"The Golden Girls",
"Florida",
"Georgia (U.S. state)"
],
"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 Golden Girls is an American movie about four older women who share a home in a state located in the Southeastern region bordered by the 24th-largest area in the U.S.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Golden Girls",
"Florida",
"Georgia (U.S. state)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2688,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2346
|
The Golden State Warriors are the fifth professional team based in San Francisco.?
|
{
"evidence": [
"Golden State Warriors",
"1956 NBA Finals"
],
"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 Golden State Warriors are the fifth professional team based in San Francisco.?\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Golden State Warriors",
"1956 NBA Finals"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2689,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1228
|
Robert John Wagner Jr. known for starring in the television shows like an Canadian mystery television series that premiered on August 25, 1979, on ABC.
|
{
"evidence": [
"Robert Wagner",
"Hart to Hart"
],
"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:\nRobert John Wagner Jr. known for starring in the television shows like an Canadian mystery television series that premiered on August 25, 1979, on ABC.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Robert Wagner",
"Hart to Hart"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2690,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1268
|
Denial is a 2016 biographical film based on Deborah Lipstadt's 2005 book History on Trial: My Day in Court with?the Denier, a?genocide of European Jews during World War II.
|
{
"evidence": [
"Denial (2016 film)",
"The Holocaust"
],
"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:\nDenial is a 2016 biographical film based on Deborah Lipstadt's 2005 book History on Trial: My Day in Court with?the Denier, a?genocide of European Jews during World War II.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Denial (2016 film)",
"The Holocaust"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2691,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8586
|
Aestheticism was first named by the critic Walter Hamilton in The Aesthetic Movement in England in 1882.
|
{
"evidence": [
"Aestheticism",
"Music"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nAestheticism was first named by the critic Walter Hamilton in The Aesthetic Movement in England in 1882.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Aestheticism",
"Music"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2692,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
34417
|
After Rossbach, Madame de Pompadour is alleged to have comforted the queen with the now-famous: "au reste, après nous, le Déluge" ("Besides, after us, the Deluge") and Italy emerged from the war diminished and virtually bankrupt.
|
{
"evidence": [
"Madame de Pompadour_sentence_55"
],
"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:\nAfter Rossbach, Madame de Pompadour is alleged to have comforted the queen with the now-famous: \"au reste, après nous, le Déluge\" (\"Besides, after us, the Deluge\") and Italy emerged from the war diminished and virtually bankrupt.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Madame de Pompadour_sentence_55"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2693,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.943",
"value": "?search="
},
{
"operation": "search",
"time": "8.801",
"value": "Madame de Pompadour"
},
{
"operation": "Now on",
"time": "10.628",
"value": "Madame de Pompadour"
},
{
"operation": "Page search",
"time": "45.629",
"value": "rossbach"
},
{
"operation": "Highlighting",
"time": "52.317",
"value": "Madame de Pompadour_sentence_55"
},
{
"operation": "finish",
"time": "68.653",
"value": "finish"
}
]
}
|
80125
|
The Goose Lake Valley is not found in Los Angeles, California.
|
{
"evidence": [
"Goose Lake Valley_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 Goose Lake Valley is not found in Los Angeles, California.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Goose Lake Valley_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2694,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.603",
"value": "undefined"
},
{
"operation": "search",
"time": "7.046",
"value": "Contains...Goose Lake Valley"
},
{
"operation": "Now on",
"time": "7.572",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "9.993",
"value": "Goose Lake Valley"
},
{
"operation": "Now on",
"time": "10.386",
"value": "Goose Lake Valley"
},
{
"operation": "Highlighting",
"time": "12.654",
"value": "Goose Lake Valley_sentence_0"
},
{
"operation": "finish",
"time": "20.166",
"value": "finish"
}
]
}
|
exfever_1271
|
Henry M. Leland founded the Lincoln Motor Company in 1917, named after the 16th president of the United States.
|
{
"evidence": [
"Lincoln Motor Company",
"Abraham Lincoln"
],
"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:\nHenry M. Leland founded the Lincoln Motor Company in 1917, named after the 16th president of the United States.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Lincoln Motor Company",
"Abraham Lincoln"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2695,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_530
|
The Tracey Fragments is a 2007 psychological drama film written by a Canadian writer and playwright.
|
{
"evidence": [
"The Tracey Fragments (film)",
"Maureen Medved"
],
"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 Tracey Fragments is a 2007 psychological drama film written by a Canadian writer and playwright.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Tracey Fragments (film)",
"Maureen Medved"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2696,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6433
|
George R. R. Martin has written the most selling book in the United States.
|
{
"evidence": [
"Bran Stark",
"A Game of Thrones"
],
"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:\nGeorge R. R. Martin has written the most selling book in the United States.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Bran Stark",
"A Game of Thrones"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2697,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
26039
|
The melaleuca cardiophylla, a plant in the myrtle family which is endemic to Western Australia, has petals with edges, that are almost transparent, white stamens, which are arranged in bundles, and almost spherical fruits.
|
{
"evidence": [
"Melaleuca cardiophylla_sentence_0",
"Melaleuca cardiophylla_sentence_6",
"Melaleuca cardiophylla_sentence_7",
"Melaleuca cardiophylla_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:\nThe melaleuca cardiophylla, a plant in the myrtle family which is endemic to Western Australia, has petals with edges, that are almost transparent, white stamens, which are arranged in bundles, and almost spherical fruits.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Melaleuca cardiophylla_sentence_0",
"Melaleuca cardiophylla_sentence_6",
"Melaleuca cardiophylla_sentence_7",
"Melaleuca cardiophylla_sentence_9"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2698,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.117",
"value": "undefined"
},
{
"operation": "search",
"time": "17.084",
"value": "melaleuca cardiophylla"
},
{
"operation": "Now on",
"time": "18.53",
"value": "Melaleuca cardiophylla"
},
{
"operation": "Highlighting",
"time": "27.342",
"value": "Melaleuca cardiophylla_sentence_0"
},
{
"operation": "Page search",
"time": "46.441",
"value": "petals"
},
{
"operation": "Highlighting",
"time": "75.295",
"value": "Melaleuca cardiophylla_sentence_6"
},
{
"operation": "Highlighting",
"time": "85.61",
"value": "Melaleuca cardiophylla_sentence_7"
},
{
"operation": "Page search",
"time": "99.949",
"value": "fruit"
},
{
"operation": "Highlighting",
"time": "114.081",
"value": "Melaleuca cardiophylla_sentence_9"
},
{
"operation": "finish",
"time": "157.238",
"value": "finish"
}
]
}
|
exfever_2631
|
Henry John Deutschendorf Jr. was an American singer-songwriter, actor, activist, and humanitarian who has been called "among the most beloved entertainers of his era" by an American online music database that catalogs more than three billion album entries and 30 million tracks, as well as information on musicians and bands.
|
{
"evidence": [
"John Denver",
"AllMusic"
],
"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:\nHenry John Deutschendorf Jr. was an American singer-songwriter, actor, activist, and humanitarian who has been called \"among the most beloved entertainers of his era\" by an American online music database that catalogs more than three billion album entries and 30 million tracks, as well as information on musicians and bands.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"John Denver",
"AllMusic"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2699,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1397
|
Wyatt Earp was an English lawman who took part in the famous thirty-second shootout between lawmen led by Virgil Earp and members of the Cowboys.
|
{
"evidence": [
"Wyatt Earp",
"Gunfight at the O.K. Corral"
],
"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:\nWyatt Earp was an English lawman who took part in the famous thirty-second shootout between lawmen led by Virgil Earp and members of the Cowboys.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Wyatt Earp",
"Gunfight at the O.K. Corral"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2700,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
38167
|
Sandy Nava, who played in 29 games his rookie season, returned to the Grays for two more seasons, continuing to be the back-up catcher to Barney Gilligan.
|
{
"evidence": [
"Sandy Nava_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:\nSandy Nava, who played in 29 games his rookie season, returned to the Grays for two more seasons, continuing to be the back-up catcher to Barney Gilligan.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Sandy Nava_sentence_5"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2701,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.096",
"value": "?search="
},
{
"operation": "search",
"time": "15.16",
"value": "Sandy Nava"
},
{
"operation": "Now on",
"time": "16.04",
"value": "Sandy Nava"
},
{
"operation": "Highlighting",
"time": "24.601",
"value": "Sandy Nava_sentence_5"
},
{
"operation": "finish",
"time": "89.205",
"value": "finish"
}
]
}
|
exfever_3865
|
Vincent Cassel is an Italian actor who garnered wide recognition with English-speaking audiences for his performances in a 2004 American heist comedy film directed by Steven Soderbergh.
|
{
"evidence": [
"Vincent Cassel",
"Ocean's Twelve"
],
"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:\nVincent Cassel is an Italian actor who garnered wide recognition with English-speaking audiences for his performances in a 2004 American heist comedy film directed by Steven Soderbergh.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Vincent Cassel",
"Ocean's Twelve"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2702,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
42349
|
The church of San Esteban in Anúcita (a village in which Spanish is the local language) has an altarpiece that was built in 1792 and an outer wall window bearing a Chrismon.
|
{
"evidence": [
"Anúcita_sentence_7"
],
"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 church of San Esteban in Anúcita (a village in which Spanish is the local language) has an altarpiece that was built in 1792 and an outer wall window bearing a Chrismon.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Anúcita_sentence_7"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2703,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.873",
"value": "undefined"
},
{
"operation": "search",
"time": "14.985",
"value": "San Esteban in Anúcita"
},
{
"operation": "Now on",
"time": "16.583",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "21.692",
"value": "Anúcita"
},
{
"operation": "Now on",
"time": "22.886",
"value": "Anúcita"
},
{
"operation": "Highlighting",
"time": "42.931",
"value": "Anúcita_sentence_7"
},
{
"operation": "Page search",
"time": "69.753",
"value": "window bearing a Chrismon"
},
{
"operation": "search",
"time": "74.09",
"value": "window bearing a Chrismon"
},
{
"operation": "Now on",
"time": "75.828",
"value": "undefined"
},
{
"operation": "search",
"time": "94.381",
"value": "altarpiece that was built in 1792"
},
{
"operation": "Now on",
"time": "95.911",
"value": "undefined"
},
{
"operation": "search",
"time": "120.853",
"value": "The church of San Esteban in Anúcita ("
},
{
"operation": "Now on",
"time": "121.378",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "125.295",
"value": "Education Anúcita_section_0"
},
{
"operation": "Now on",
"time": "125.69",
"value": "Anúcita#Education An.C3.BAcita section 0"
},
{
"operation": "search",
"time": "137.128",
"value": "alterpice"
},
{
"operation": "Now on",
"time": "137.882",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "144.677",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "144.765",
"value": "Anúcita#Education An.C3.BAcita section 0"
},
{
"operation": "finish",
"time": "225.371",
"value": "finish"
}
]
}
|
exfever_924
|
Leonardo di ser Piero da Vinci was an Irish polymath of the High Renaissance whose collective works composed a contribution to later generations of artists matched only by an Italian sculptor, painter, architect, and poet of the High Renaissance born in a medieval and early modern state that was centered in the Italian city of Florence in Tuscany.
|
{
"evidence": [
"Leonardo da Vinci",
"Michelangelo",
"Republic of Florence"
],
"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:\nLeonardo di ser Piero da Vinci was an Irish polymath of the High Renaissance whose collective works composed a contribution to later generations of artists matched only by an Italian sculptor, painter, architect, and poet of the High Renaissance born in a medieval and early modern state that was centered in the Italian city of Florence in Tuscany.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Leonardo da Vinci",
"Michelangelo",
"Republic of Florence"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2704,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6801
|
Michael Keaton has won an Academy Award for his role in The Other Guys.
|
{
"evidence": [
"Michael Keaton",
"DC Comics"
],
"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:\nMichael Keaton has won an Academy Award for his role in The Other Guys.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Michael Keaton",
"DC Comics"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2705,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
66407
|
Escape from Monster Manor is a drama film.
|
{
"evidence": [
"Escape from Monster Manor_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:\nEscape from Monster Manor is a drama film.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Escape from Monster Manor_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2706,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.828",
"value": "?search="
},
{
"operation": "search",
"time": "12.58",
"value": "Escape from Monster Manor"
},
{
"operation": "Now on",
"time": "20.951",
"value": "Escape from Monster Manor"
},
{
"operation": "Highlighting",
"time": "29.596",
"value": "Escape from Monster Manor_sentence_0"
},
{
"operation": "search",
"time": "38.346",
"value": "Contains...Escape from Monster Manor"
},
{
"operation": "Now on",
"time": "42.069",
"value": "undefined"
},
{
"operation": "finish",
"time": "67.236",
"value": "finish"
}
]
}
|
exfever_787
|
The Academy Awards are awards for artistic and technical merit in the film industry that was originally sculpted by George Stanley from a design sketch by Cedric Gibbons which first presented it in 1930 at a private dinner hosted by Douglas Fairbanks in a historic hotel located at 7000 Hollywood Boulevard in the Hollywood district of Los Angeles, California.
|
{
"evidence": [
"Academy Awards",
"The Hollywood Roosevelt Hotel"
],
"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 Academy Awards are awards for artistic and technical merit in the film industry that was originally sculpted by George Stanley from a design sketch by Cedric Gibbons which first presented it in 1930 at a private dinner hosted by Douglas Fairbanks in a historic hotel located at 7000 Hollywood Boulevard in the Hollywood district of Los Angeles, California.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Academy Awards",
"The Hollywood Roosevelt Hotel"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2707,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5994
|
Bring It On is a 2000 American teen cheerleading comedy film directed by an American television and film director with a 2008 romantic comedy film credited to him.
|
{
"evidence": [
"Bring It On (film)",
"Peyton Reed",
"Yes Man (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:\nBring It On is a 2000 American teen cheerleading comedy film directed by an American television and film director with a 2008 romantic comedy film credited to him.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Bring It On (film)",
"Peyton Reed",
"Yes Man (film)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2708,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
49625
|
THe 2019–21 ICC World Test Championship, the inaugural edition of the ICC World Test Championship of Test cricket, is distributed into four matches in which a players must earn 24 to 60 points for a win.
|
{
"evidence": [
"2019–21 ICC World Test Championship_sentence_4"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nTHe 2019–21 ICC World Test Championship, the inaugural edition of the ICC World Test Championship of Test cricket, is distributed into four matches in which a players must earn 24 to 60 points for a win.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"2019–21 ICC World Test Championship_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2709,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.964",
"value": "?search="
},
{
"operation": "search",
"time": "8.923",
"value": "2019–21 ICC World Test Championship"
},
{
"operation": "Now on",
"time": "13.731",
"value": "2019–21 ICC World Test Championship"
},
{
"operation": "Highlighting",
"time": "22.729",
"value": "2019–21 ICC World Test Championship_sentence_0"
},
{
"operation": "Page search",
"time": "198.138",
"value": "four matches"
},
{
"operation": "Highlighting deleted",
"time": "214.948",
"value": "2019–21 ICC World Test Championship_sentence_0"
},
{
"operation": "Highlighting",
"time": "218.701",
"value": "2019–21 ICC World Test Championship_sentence_4"
},
{
"operation": "finish",
"time": "242.87",
"value": "finish"
}
]
}
|
exfever_8460
|
Shakira is a Colombian singer who rose to prominence in Hispanic countries with her third studio album, released by Sony Music and Warner Bros.
|
{
"evidence": [
"Shakira",
"Pies Descalzos"
],
"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:\nShakira is a Colombian singer who rose to prominence in Hispanic countries with her third studio album, released by Sony Music and Warner Bros.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Shakira",
"Pies Descalzos"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2710,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6722
|
Dilwale Dulhania Le Jayenge is an Indian Hindi-language musical film that involves romantic love stories that typically follow the journey of two people exploring whether they are romantically or sexually compatible by participating in dates with each other.
|
{
"evidence": [
"Dilwale Dulhania Le Jayenge",
"Romance film",
"Dating"
],
"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:\nDilwale Dulhania Le Jayenge is an Indian Hindi-language musical film that involves romantic love stories that typically follow the journey of two people exploring whether they are romantically or sexually compatible by participating in dates with each other.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Dilwale Dulhania Le Jayenge",
"Romance film",
"Dating"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2711,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
47677
|
Teh 2016 Pennsylvania Auditor General election was held on November 8, 2016 to elect a new Pennsylvania Attorney General which was won by John Sweeney of the Democatric party with a total vote of 2,958,818 .
|
{
"evidence": [
"2016 Pennsylvania Auditor General election_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:\nTeh 2016 Pennsylvania Auditor General election was held on November 8, 2016 to elect a new Pennsylvania Attorney General which was won by John Sweeney of the Democatric party with a total vote of 2,958,818\t.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"2016 Pennsylvania Auditor General election_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2712,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.076",
"value": "?search="
},
{
"operation": "search",
"time": "18.004",
"value": "2016 Pennsylvania Auditor General election"
},
{
"operation": "Now on",
"time": "18.88",
"value": "2016 Pennsylvania Auditor General election"
},
{
"operation": "Highlighting",
"time": "36.055",
"value": "2016 Pennsylvania Auditor General election_sentence_0"
},
{
"operation": "Page search",
"time": "39.342",
"value": "Pennsylvania Attorney General"
},
{
"operation": "Page search",
"time": "39.501",
"value": "Pennsylvania Attorney General"
},
{
"operation": "finish",
"time": "66.962",
"value": "finish"
}
]
}
|
72193
|
Blanche's husband Harry Morton (Larry Keating) was George's accountant and also present were Harry von Zell, Ronnie Burns, and Judi Meredith, all playing themselves.
|
{
"evidence": [
"The George Burns Show_sentence_6",
"The George Burns Show_sentence_7"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBlanche's husband Harry Morton (Larry Keating) was George's accountant and also present were Harry von Zell, Ronnie Burns, and Judi Meredith, all playing themselves.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"The George Burns Show_sentence_6",
"The George Burns Show_sentence_7"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2713,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.655",
"value": "?search="
},
{
"operation": "search",
"time": "25.628",
"value": "Harry Morton"
},
{
"operation": "Now on",
"time": "26.755",
"value": "Harry Morton"
},
{
"operation": "hyperlink",
"time": "34.128",
"value": "Harry Morton (disambiguation)"
},
{
"operation": "Now on",
"time": "34.696",
"value": "Harry Morton (disambiguation)"
},
{
"operation": "back-button-clicked",
"time": "43.213",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "43.271",
"value": "Harry Morton"
},
{
"operation": "search",
"time": "48.274",
"value": "Harry Morton character"
},
{
"operation": "Now on",
"time": "49.854",
"value": "undefined"
},
{
"operation": "search",
"time": "85.848",
"value": "blanche morton"
},
{
"operation": "Now on",
"time": "87.508",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "98.025",
"value": "Brighton Beach Memoirs (film)"
},
{
"operation": "Now on",
"time": "98.349",
"value": "Brighton Beach Memoirs (film)"
},
{
"operation": "Page search",
"time": "110.897",
"value": "George's accountan"
},
{
"operation": "Page search",
"time": "112.199",
"value": "George's accountant"
},
{
"operation": "back-button-clicked",
"time": "120.6",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "121.087",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "122.991",
"value": "Larry Keating"
},
{
"operation": "Now on",
"time": "123.847",
"value": "Larry Keating"
},
{
"operation": "hyperlink",
"time": "152.396",
"value": "The George Burns and Gracie Allen Show"
},
{
"operation": "Now on",
"time": "153.526",
"value": "The George Burns and Gracie Allen Show"
},
{
"operation": "hyperlink",
"time": "171.982",
"value": "List of The George Burns and Gracie Allen Show episodes"
},
{
"operation": "Now on",
"time": "180.172",
"value": "List of The George Burns and Gracie Allen Show episodes"
},
{
"operation": "search",
"time": "266.518",
"value": "Gracie interferes after Harry Von Zell goes out with a married woman"
},
{
"operation": "Now on",
"time": "291.883",
"value": "undefined"
},
{
"operation": "search",
"time": "311.175",
"value": "He played the hesitant, somewhat bewildered friend"
},
{
"operation": "Now on",
"time": "312.884",
"value": "undefined"
},
{
"operation": "search",
"time": "353.967",
"value": "harry von zell"
},
{
"operation": "Now on",
"time": "354.868",
"value": "Harry von Zell"
},
{
"operation": "search",
"time": "392.91",
"value": "(Larry Keating) was George's accountant"
},
{
"operation": "Now on",
"time": "394.468",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "398.528",
"value": "Overview The George Burns Show_section_0"
},
{
"operation": "Now on",
"time": "399.084",
"value": "The George Burns Show#Overview The George Burns Show section 0"
},
{
"operation": "Highlighting",
"time": "401.545",
"value": "The George Burns Show_sentence_6"
},
{
"operation": "Highlighting",
"time": "407.811",
"value": "The George Burns Show_sentence_7"
},
{
"operation": "finish",
"time": "432.408",
"value": "finish"
}
]
}
|
exfever_1736
|
The 66th Primetime Emmy Awards was held on August 25, 2014, at the Hollywood Theatre in Downtown Los Angeles, California, and was broadcast on the Comcast-owned commercial broadcast television and radio network in the United States.
|
{
"evidence": [
"66th Primetime Emmy Awards",
"NBC"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe 66th Primetime Emmy Awards was held on August 25, 2014, at the Hollywood Theatre in Downtown Los Angeles, California, and was broadcast on the Comcast-owned commercial broadcast television and radio network in the United States.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"66th Primetime Emmy Awards",
"NBC"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2714,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6569
|
A Game of Thrones is the first novel in A Song of Ice and Fire that was nominated for the set of awards given each year for the best fantasy fiction published during the previous calendar month.
|
{
"evidence": [
"A Game of Thrones",
"World Fantasy Award"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nA Game of Thrones is the first novel in A Song of Ice and Fire that was nominated for the set of awards given each year for the best fantasy fiction published during the previous calendar month.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"A Game of Thrones",
"World Fantasy Award"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2715,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2227
|
Entourage is a 2015 American comedy film written, directed and produced by an American podcaster, screenwriter and film and TV director, who attended Tulane University.
|
{
"evidence": [
"Entourage (film)",
"Doug Ellin"
],
"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:\nEntourage is a 2015 American comedy film written, directed and produced by an American podcaster, screenwriter and film and TV director, who attended Tulane University.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Entourage (film)",
"Doug Ellin"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2716,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7048
|
Billie Jean King is an American former world No. 1 badminton player who founded the principal organizing body of women's professional tennis, which governs the worldwide professional tennis tour for women.?
|
{
"evidence": [
"Billie Jean King",
"Women's Tennis Association"
],
"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:\nBillie Jean King is an American former world No. 1 badminton player who founded the principal organizing body of women's professional tennis, which governs the worldwide professional tennis tour for women.?\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Billie Jean King",
"Women's Tennis Association"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2717,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_939
|
The great white shark is a species of large Lamniformes that can be found in the coastal surface waters of all the major oceans.
|
{
"evidence": [
"Great white shark",
"Lamniformes"
],
"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 great white shark is a species of large Lamniformes that can be found in the coastal surface waters of all the major oceans.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Great white shark",
"Lamniformes"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2718,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_543
|
Tobey Maguire is an American actor known for his role in a 2003 American horror film directed by Gary Ross that is based on the best-selling 1999 non-fiction book by Laura Hillenbrand.
|
{
"evidence": [
"Tobey Maguire",
"Seabiscuit (film)"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nTobey Maguire is an American actor known for his role in a 2003 American horror film directed by Gary Ross that is based on the best-selling 1999 non-fiction book by Laura Hillenbrand.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tobey Maguire",
"Seabiscuit (film)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2719,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
43056
|
In his sophomore year, Matt Costello accomplished 43 blocks, which were the best total in school history, and he also ranked second on the team in field goal percentage even though he missed four games due to mononucleosis.
|
{
"evidence": [
"Matt Costello_sentence_27"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nIn his sophomore year, Matt Costello accomplished 43 blocks, which were the best total in school history, and he also ranked second on the team in field goal percentage even though he missed four games due to mononucleosis.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Matt Costello_sentence_27"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2720,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.13",
"value": "?search="
},
{
"operation": "search",
"time": "23.191",
"value": "Matt Costello (footballer)"
},
{
"operation": "Now on",
"time": "24.352",
"value": "Matt Costello (footballer)"
},
{
"operation": "search",
"time": "32.67",
"value": "Matt Costello"
},
{
"operation": "Now on",
"time": "33.867",
"value": "Matt Costello"
},
{
"operation": "hyperlink",
"time": "42.708",
"value": "2.2 Sophomore year Matt Costello_section_3"
},
{
"operation": "Highlighting",
"time": "62.264",
"value": "Matt Costello_sentence_27"
},
{
"operation": "finish",
"time": "74.213",
"value": "finish"
}
]
}
|
exfever_4925
|
Beautiful is a 2000 comedy-drama film starring an American former child actress, best known for being "The Pepsi Girl".
|
{
"evidence": [
"Beautiful (2000 film)",
"Hallie Eisenberg"
],
"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:\nBeautiful is a 2000 comedy-drama film starring an American former child actress, best known for being \"The Pepsi Girl\".\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Beautiful (2000 film)",
"Hallie Eisenberg"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2721,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
94688
|
Zinaida Botschantzeva's career was only in private industry.
|
{
"evidence": [
"Zinaida Botschantzeva_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:\nZinaida Botschantzeva's career was only in private industry.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Zinaida Botschantzeva_sentence_7"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2722,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.771",
"value": "?search="
},
{
"operation": "search",
"time": "12.653",
"value": "Zinaida Botschantzeva"
},
{
"operation": "Now on",
"time": "13.548",
"value": "Zinaida Botschantzeva"
},
{
"operation": "Highlighting",
"time": "62.946",
"value": "Zinaida Botschantzeva_sentence_7"
},
{
"operation": "hyperlink",
"time": "91.561",
"value": "Botanical Garden"
},
{
"operation": "Now on",
"time": "94.297",
"value": "Botanical garden"
},
{
"operation": "finish",
"time": "105.281",
"value": "finish"
}
]
}
|
26834
|
During Robert Oakeshott's Oxford years in 1956, an anti-Soviet uprising broke out in Hungary and Oakeshott and a fellow student made a goodwill visit to Budapest, taking a consignment of medical supplies.
|
{
"evidence": [
"Robert Oakeshott_sentence_11",
"Robert Oakeshott_sentence_10"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nDuring Robert Oakeshott's Oxford years in 1956, an anti-Soviet uprising broke out in Hungary and Oakeshott and a fellow student made a goodwill visit to Budapest, taking a consignment of medical supplies.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Robert Oakeshott_sentence_11",
"Robert Oakeshott_sentence_10"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2723,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.035",
"value": "?search="
},
{
"operation": "search",
"time": "22.551",
"value": "Robert Oakeshott"
},
{
"operation": "Now on",
"time": "23.227",
"value": "Robert Oakeshott"
},
{
"operation": "Highlighting",
"time": "83.932",
"value": "Robert Oakeshott_sentence_11"
},
{
"operation": "Highlighting",
"time": "108.613",
"value": "Robert Oakeshott_sentence_10"
},
{
"operation": "finish",
"time": "130.659",
"value": "finish"
}
]
}
|
exfever_4011
|
The United States Congress was created by the Constitution of the United States to replace its legislative function as the governing body of the United States of America.
|
{
"evidence": [
"United States Congress",
"Congress of the Confederation"
],
"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 United States Congress was created by the Constitution of the United States to replace its legislative function as the governing body of the United States of America.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"United States Congress",
"Congress of the Confederation"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2724,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
34943
|
Timed event system DEVS (Discrete Event System Specification), a formalism for modeling and analyzing general systems, was invented by Bernard P. Zeigler published in his second book released in 1976.
|
{
"evidence": [
"DEVS_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:\nTimed event system DEVS (Discrete Event System Specification), a formalism for modeling and analyzing general systems, was invented by Bernard P. Zeigler published in his second book released in 1976.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"DEVS_sentence_5"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2725,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.33",
"value": "?search="
},
{
"operation": "search",
"time": "9.859",
"value": "DEVS"
},
{
"operation": "Now on",
"time": "10.755",
"value": "DEVS"
},
{
"operation": "Highlighting",
"time": "13.949",
"value": "DEVS_sentence_1"
},
{
"operation": "Highlighting",
"time": "31.809",
"value": "DEVS_sentence_2"
},
{
"operation": "Highlighting",
"time": "41.656",
"value": "DEVS_sentence_5"
},
{
"operation": "Highlighting",
"time": "49.38",
"value": "DEVS_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "63.442",
"value": "DEVS_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "72.66",
"value": "DEVS_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "73.41",
"value": "DEVS_sentence_1"
},
{
"operation": "finish",
"time": "86.615",
"value": "finish"
}
]
}
|
exfever_7366
|
Pablo Picasso was a Spanish painter whose most famous works are the dramatic portrayals of the aerial bombing of the Basque town of Guernica during the Spanish Civil War.
|
{
"evidence": [
"Pablo Picasso",
"Bombing of Guernica"
],
"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:\nPablo Picasso was a Spanish painter whose most famous works are the dramatic portrayals of the aerial bombing of the Basque town of Guernica during the Spanish Civil War.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Pablo Picasso",
"Bombing of Guernica"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2726,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
48311
|
The song "Sugar" by Brockhampton ( an American hip-hop musical group) didn't make it onto the top charts in 2019.
|
{
"evidence": [
"Sugar (Brockhampton song)_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:\nThe song \"Sugar\" by Brockhampton ( an American hip-hop musical group) didn't make it onto the top charts in 2019.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Sugar (Brockhampton song)_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2727,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.991",
"value": "?search="
},
{
"operation": "search",
"time": "14.096",
"value": "Sugar Brockhampton"
},
{
"operation": "Now on",
"time": "15.25",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "18.991",
"value": "Sugar (Brockhampton song)"
},
{
"operation": "Now on",
"time": "19.651",
"value": "Sugar (Brockhampton song)"
},
{
"operation": "Highlighting",
"time": "21.387",
"value": "Sugar (Brockhampton song)_sentence_0"
},
{
"operation": "hyperlink",
"time": "32.983",
"value": "Brockhampton"
},
{
"operation": "Now on",
"time": "33.648",
"value": "Brockhampton (band)"
},
{
"operation": "back-button-clicked",
"time": "40.914",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "40.965",
"value": "Sugar (Brockhampton song)"
},
{
"operation": "hyperlink",
"time": "164.418",
"value": "Ginger"
},
{
"operation": "Now on",
"time": "165.091",
"value": "Ginger (Brockhampton album)"
},
{
"operation": "Page search",
"time": "224.366",
"value": "sugar"
},
{
"operation": "Highlighting",
"time": "238.012",
"value": "Ginger (Brockhampton album)_sentence_55"
},
{
"operation": "Highlighting",
"time": "238.012",
"value": "Ginger (Brockhampton album)_title"
},
{
"operation": "Highlighting",
"time": "238.014",
"value": "Ginger (Brockhampton album)_section_3"
},
{
"operation": "Highlighting",
"time": "238.589",
"value": "Ginger (Brockhampton album)_sentence_56"
},
{
"operation": "Highlighting deleted",
"time": "258.475",
"value": "Ginger (Brockhampton album)_sentence_56"
},
{
"operation": "Highlighting deleted",
"time": "412.049",
"value": "Ginger (Brockhampton album)_sentence_55"
},
{
"operation": "Highlighting deleted",
"time": "412.051",
"value": "Ginger (Brockhampton album)_title"
},
{
"operation": "Highlighting deleted",
"time": "412.053",
"value": "Ginger (Brockhampton album)_section_3"
},
{
"operation": "back-button-clicked",
"time": "491.14",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "491.198",
"value": "Sugar (Brockhampton song)"
},
{
"operation": "finish",
"time": "521.589",
"value": "finish"
}
]
}
|
48318
|
"Be Okay" was made available for purchase on Spotify on September 19, 2013, then made it onto Billboard's Adult Top 40 chart where it reached number 25.
|
{
"evidence": [
"Be Okay_sentence_14",
"Be Okay_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:\n\"Be Okay\" was made available for purchase on Spotify on September 19, 2013, then made it onto Billboard's Adult Top 40 chart where it reached number 25.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Be Okay_sentence_14",
"Be Okay_sentence_12"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2728,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.249",
"value": "?search="
},
{
"operation": "search",
"time": "18.016",
"value": "Be Okay"
},
{
"operation": "Now on",
"time": "19.317",
"value": "Be Okay"
},
{
"operation": "Page search",
"time": "33.929",
"value": "spoti"
},
{
"operation": "Page search",
"time": "80.047",
"value": "2013"
},
{
"operation": "Highlighting",
"time": "105.46",
"value": "Be Okay_sentence_14"
},
{
"operation": "Highlighting",
"time": "120.527",
"value": "Be Okay_sentence_12"
},
{
"operation": "search",
"time": "147.211",
"value": "\"Be Okay\" \"spotify\""
},
{
"operation": "Now on",
"time": "148.592",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "159.626",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "159.68",
"value": "Be Okay"
},
{
"operation": "search",
"time": "182.627",
"value": "\"Be Okay\" \"collins\" \"spotify\""
},
{
"operation": "Now on",
"time": "183.442",
"value": "undefined"
},
{
"operation": "finish",
"time": "190.337",
"value": "finish"
}
]
}
|
20687
|
Marshall Orme Wilson was the friends to the eldest son to Richard Thornton Wilson Sr (a multimillionaire investment banker) and his family was nicknamed the Marrying Wilsons.
|
{
"evidence": [
"Marshall Orme Wilson_sentence_2",
"Richard Thornton Wilson_sentence_0",
"Marshall Orme Wilson_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:\nMarshall Orme Wilson was the friends to the eldest son to Richard Thornton Wilson Sr (a multimillionaire investment banker) and his family was nicknamed the Marrying Wilsons.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Marshall Orme Wilson_sentence_2",
"Richard Thornton Wilson_sentence_0",
"Marshall Orme Wilson_sentence_5"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2729,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.091",
"value": "?search="
},
{
"operation": "search",
"time": "11.788",
"value": "Marshall Orme Wilson"
},
{
"operation": "Now on",
"time": "14.119",
"value": "Marshall Orme Wilson"
},
{
"operation": "Highlighting",
"time": "36.605",
"value": "Marshall Orme Wilson_sentence_2"
},
{
"operation": "hyperlink",
"time": "61.285",
"value": "Richard Thornton Wilson Sr."
},
{
"operation": "Now on",
"time": "65.352",
"value": "Richard Thornton Wilson"
},
{
"operation": "Highlighting",
"time": "66.653",
"value": "Richard Thornton Wilson_sentence_0"
},
{
"operation": "back-button-clicked",
"time": "73.374",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "73.466",
"value": "Marshall Orme Wilson"
},
{
"operation": "Highlighting",
"time": "83.133",
"value": "Marshall Orme Wilson_sentence_5"
},
{
"operation": "finish",
"time": "101.963",
"value": "finish"
}
]
}
|
65355
|
In 1845, the Wallonian (a group native to Belgium) Jacques Piedboeuf, together with Hugo Jakob Talbot, Johann Leonhard Neuman and Theodor Esser, founded the steelworks OHG Piedboeuf & Co, Aachener Walz- und Hammerwerk; taken over in 1851 by Carl Ruëtz, which then continued as the Kommanditgesellschaft Carl Ruëtz & Co – Aachener Hütten-Aktien-Verein Rothe Erde.
|
{
"evidence": [
"Walloons_sentence_0",
"Rothe Erde_sentence_4",
"Rothe Erde_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:\nIn 1845, the Wallonian (a group native to Belgium) Jacques Piedboeuf, together with Hugo Jakob Talbot, Johann Leonhard Neuman and Theodor Esser, founded the steelworks OHG Piedboeuf & Co, Aachener Walz- und Hammerwerk; taken over in 1851 by Carl Ruëtz, which then continued as the Kommanditgesellschaft Carl Ruëtz & Co – Aachener Hütten-Aktien-Verein Rothe Erde.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Walloons_sentence_0",
"Rothe Erde_sentence_4",
"Rothe Erde_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2730,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.492",
"value": "?search="
},
{
"operation": "search",
"time": "29.287",
"value": "Wallonian"
},
{
"operation": "Now on",
"time": "30.448",
"value": "Walloons"
},
{
"operation": "Highlighting",
"time": "39.962",
"value": "Walloons_sentence_0"
},
{
"operation": "search",
"time": "63.047",
"value": "Contains...Jacques Piedboeuf"
},
{
"operation": "Now on",
"time": "64.008",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "72.873",
"value": "Rothe Erde"
},
{
"operation": "Now on",
"time": "75.875",
"value": "Rothe Erde"
},
{
"operation": "Highlighting",
"time": "79.545",
"value": "Rothe Erde_sentence_4"
},
{
"operation": "Highlighting",
"time": "100.792",
"value": "Rothe Erde_sentence_5"
},
{
"operation": "finish",
"time": "117.696",
"value": "finish"
}
]
}
|
16897
|
Bob Wiese is the all-time top scorer of the Michigan Wolverines football team.
|
{
"evidence": [
"Bob Wiese_sentence_9"
],
"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:\nBob Wiese is the all-time top scorer of the Michigan Wolverines football team.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Bob Wiese_sentence_9"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2731,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.975",
"value": "undefined"
},
{
"operation": "search",
"time": "9.984",
"value": "Bob Wiese"
},
{
"operation": "Now on",
"time": "10.985",
"value": "Bob Wiese"
},
{
"operation": "Page search",
"time": "25.314",
"value": "score"
},
{
"operation": "Highlighting",
"time": "40.233",
"value": "Bob Wiese_sentence_9"
},
{
"operation": "finish",
"time": "55.908",
"value": "finish"
}
]
}
|
29054
|
Written in "clear, concise language" and the oldest illustrated Yiddish-language book, the Paris Book of Customs came into the possession of Cardinal Richelieu some time after 1632.
|
{
"evidence": [
"Paris Book of Customs_sentence_0",
"Paris Book of Customs_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:\nWritten in \"clear, concise language\" and the oldest illustrated Yiddish-language book, the Paris Book of Customs came into the possession of Cardinal Richelieu some time after 1632.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Paris Book of Customs_sentence_0",
"Paris Book of Customs_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2732,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.807",
"value": "?search="
},
{
"operation": "search",
"time": "16.853",
"value": "Paris Book of Customs"
},
{
"operation": "Now on",
"time": "17.75",
"value": "Paris Book of Customs"
},
{
"operation": "Highlighting",
"time": "21.864",
"value": "Paris Book of Customs_sentence_0"
},
{
"operation": "Highlighting",
"time": "32.19",
"value": "Paris Book of Customs_sentence_3"
},
{
"operation": "finish",
"time": "56.288",
"value": "finish"
}
]
}
|
exfever_7555
|
Michael Phelps already tied the record of eight medals of any color at a single Games at the international multi-sport event since the first modern incarnation held in Athens, Greece.
|
{
"evidence": [
"Michael Phelps",
"2004 Summer Olympics",
"1896 Summer Olympics"
],
"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:\nMichael Phelps already tied the record of eight medals of any color at a single Games at the international multi-sport event since the first modern incarnation held in Athens, Greece.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Michael Phelps",
"2004 Summer Olympics",
"1896 Summer Olympics"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2733,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7647
|
Christopher O'Dowd is an Indian actor and comedian who created and starred in Moone Boy by a British pay television channel that broadcast exclusively as a broadcaster and telecommunications company that provides television and broadband Internet services, fixed-line and mobile telephone services.
|
{
"evidence": [
"Chris O'Dowd",
"Sky 1",
"Sky UK"
],
"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:\nChristopher O'Dowd is an Indian actor and comedian who created and starred in Moone Boy by a British pay television channel that broadcast exclusively as a broadcaster and telecommunications company that provides television and broadband Internet services, fixed-line and mobile telephone services.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Chris O'Dowd",
"Sky 1",
"Sky UK"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2734,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4907
|
Chris Evans, a Canadian actor, gained attention for his portrayal of the character The Human Torch in the Fantastic Four from the Publishing Company, the brand name and primary imprint of Marvel Worldwide, Inc.
|
{
"evidence": [
"Chris Evans (actor)",
"Marvel Comics"
],
"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:\nChris Evans, a Canadian actor, gained attention for his portrayal of the character The Human Torch in the Fantastic Four from the Publishing Company, the brand name and primary imprint of Marvel Worldwide, Inc.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Chris Evans (actor)",
"Marvel Comics"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2735,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7101
|
George Harrison was an English musician, singer-songwriter, and music and film producer who achieved international fame as the lead guitarist of the English rock band formed in 1961 in a city and metropolitan borough in Merseyside, England.
|
{
"evidence": [
"George Harrison",
"The Beatles",
"Liverpool"
],
"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:\nGeorge Harrison was an English musician, singer-songwriter, and music and film producer who achieved international fame as the lead guitarist of the English rock band formed in 1961 in a city and metropolitan borough in Merseyside, England.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"George Harrison",
"The Beatles",
"Liverpool"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2736,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3597
|
Courtney Michelle Love is a British singer who is the lead vocalist of an American alternative rock band formed in Los Angeles, California in 1989.
|
{
"evidence": [
"Courtney Love",
"Hole (band)"
],
"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:\nCourtney Michelle Love is a British singer who is the lead vocalist of an American alternative rock band formed in Los Angeles, California in 1989.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Courtney Love",
"Hole (band)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2737,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
72252
|
William Dering is first recorded as a dancing master in Philadelphia, he is said to have been respected as well for his talents on the French horn and in 1737 he moved to Gloucester County.
|
{
"evidence": [
"William Dering_sentence_2",
"William Dering_sentence_3",
"William Dering_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:\nWilliam Dering is first recorded as a dancing master in Philadelphia, he is said to have been respected as well for his talents on the French horn and in 1737 he moved to Gloucester County.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"William Dering_sentence_2",
"William Dering_sentence_3",
"William Dering_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2738,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.679",
"value": "?search="
},
{
"operation": "search",
"time": "32.32",
"value": "William Dering"
},
{
"operation": "Now on",
"time": "32.85",
"value": "William Dering"
},
{
"operation": "Highlighting",
"time": "50.199",
"value": "William Dering_sentence_2"
},
{
"operation": "Highlighting",
"time": "62.574",
"value": "William Dering_sentence_3"
},
{
"operation": "Highlighting",
"time": "71.014",
"value": "William Dering_sentence_5"
},
{
"operation": "hyperlink",
"time": "76.621",
"value": "Gloucester County, Virginia"
},
{
"operation": "Now on",
"time": "79.548",
"value": "Gloucester County, Virginia"
},
{
"operation": "finish",
"time": "185.781",
"value": "finish"
}
]
}
|
exfever_8034
|
The Right Stuff is a 1983 American epic historical drama film written and directed by an American film director and screenwriter who has directed fifteen films including a 1979 American comedy-drama film starring Ken Wahl.
|
{
"evidence": [
"The Right Stuff (film)",
"Philip Kaufman",
"The Wanderers (1979 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:\nThe Right Stuff is a 1983 American epic historical drama film written and directed by an American film director and screenwriter who has directed fifteen films including a 1979 American comedy-drama film starring Ken Wahl.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Right Stuff (film)",
"Philip Kaufman",
"The Wanderers (1979 film)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2739,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
66881
|
The Sacramento Surge was the third American team to win a World Bowl.
|
{
"evidence": [
"Sacramento Surge_sentence_7"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Sacramento Surge was the third American team to win a World Bowl.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Sacramento Surge_sentence_7"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2740,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "44.629",
"value": "undefined"
},
{
"operation": "search",
"time": "54.748",
"value": "Sacramento Surge"
},
{
"operation": "Now on",
"time": "57.245",
"value": "Sacramento Surge"
},
{
"operation": "Highlighting",
"time": "77.701",
"value": "Sacramento Surge_sentence_7"
},
{
"operation": "finish",
"time": "137.73",
"value": "finish"
}
]
}
|
51530
|
The Youth Parliament was established by Ishita Chaudhry and legally incorporated as 'The YP Foundation' in order to become a for-profit organization.
|
{
"evidence": [
"The Youth Parliament_sentence_2",
"The Youth Parliament_sentence_3"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Youth Parliament was established by Ishita Chaudhry and legally incorporated as 'The YP Foundation' in order to become a for-profit organization.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"The Youth Parliament_sentence_2",
"The Youth Parliament_sentence_3"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2741,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.001",
"value": "?search="
},
{
"operation": "search",
"time": "0.013",
"value": "Youth Parliament"
},
{
"operation": "Now on",
"time": "0.014",
"value": "List of youth parliaments"
},
{
"operation": "search",
"time": "0.06",
"value": "The Youth Parliament was established by Ishita Chaudhry and legally incorporated as 'The YP Foundation'"
},
{
"operation": "Now on",
"time": "0.062",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "0.067",
"value": "The Youth Parliament"
},
{
"operation": "Now on",
"time": "0.067",
"value": "The Youth Parliament"
},
{
"operation": "Highlighting",
"time": "0.074",
"value": "The Youth Parliament_sentence_1"
},
{
"operation": "Highlighting",
"time": "0.079",
"value": "The Youth Parliament_sentence_2"
},
{
"operation": "Highlighting",
"time": "0.141",
"value": "The Youth Parliament_sentence_3"
},
{
"operation": "Page search",
"time": "0.154",
"value": "profit"
},
{
"operation": "Highlighting deleted",
"time": "0.163",
"value": "The Youth Parliament_sentence_1"
},
{
"operation": "finish",
"time": "0.194",
"value": "finish"
},
{
"operation": "Now on",
"time": "1621149598.565",
"value": "The Youth Parliament"
},
{
"operation": "search",
"time": "1621149607.707",
"value": "Youth Parliament"
},
{
"operation": "Now on",
"time": "1621149608.153",
"value": "Youth Parliament"
},
{
"operation": "finish",
"time": "1621149635.697",
"value": "finish"
}
]
}
|
90955
|
The Chiman Saparaiya minister had a portfolio in agriculture as well as energy, after the first sworn minister Vijay Rupani in 2016.
|
{
"evidence": [
"Chiman Saparaiya_sentence_0",
"Chiman Saparaiya_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:\nThe Chiman Saparaiya minister had a portfolio in agriculture as well as energy, after the first sworn minister Vijay Rupani in 2016.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Chiman Saparaiya_sentence_0",
"Chiman Saparaiya_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2742,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.609",
"value": "?search="
},
{
"operation": "search",
"time": "39.387",
"value": "The Chiman Saparaiya minister"
},
{
"operation": "Now on",
"time": "40.315",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "57.262",
"value": "Chiman Saparaiya"
},
{
"operation": "Now on",
"time": "58.363",
"value": "Chiman Saparaiya"
},
{
"operation": "Highlighting",
"time": "68.678",
"value": "Chiman Saparaiya_sentence_0"
},
{
"operation": "Highlighting",
"time": "71.358",
"value": "Chiman Saparaiya_sentence_1"
},
{
"operation": "finish",
"time": "164.863",
"value": "finish"
}
]
}
|
49966
|
Elmar Hess, a German artist who works on film and video art, photography, installation and objects, created a mastermind entitled "War Years," in which he styled a relationship crisis to seem like World War II, a huge kitchen table turns into a battlefield on which quarreling lovers are staged to represent the meeting between Winston Churchill and Adolf Hitler.
|
{
"evidence": [
"Elmar Hess_sentence_1",
"Elmar Hess_sentence_12",
"Elmar Hess_sentence_13"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nElmar Hess, a German artist who works on film and video art, photography, installation and objects, created a mastermind entitled \"War Years,\" in which he styled a relationship crisis to seem like World War II, a huge kitchen table turns into a battlefield on which quarreling lovers are staged to represent the meeting between Winston Churchill and Adolf Hitler.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Elmar Hess_sentence_1",
"Elmar Hess_sentence_12",
"Elmar Hess_sentence_13"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2743,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.801",
"value": "?search="
},
{
"operation": "search",
"time": "27.121",
"value": "Elmar Hess"
},
{
"operation": "Now on",
"time": "27.846",
"value": "Elmar Hess"
},
{
"operation": "Highlighting",
"time": "33.801",
"value": "Elmar Hess_sentence_1"
},
{
"operation": "Page search",
"time": "45.339",
"value": "war years"
},
{
"operation": "hyperlink",
"time": "50.994",
"value": "2.1 \"War Years\" Elmar Hess_section_2"
},
{
"operation": "Highlighting",
"time": "69.963",
"value": "Elmar Hess_sentence_12"
},
{
"operation": "Highlighting",
"time": "71.365",
"value": "Elmar Hess_sentence_13"
},
{
"operation": "finish",
"time": "92.766",
"value": "finish"
}
]
}
|
81359
|
Unable to find any bars selling beer, the Stooges opt to become bootleggers and brew some of the stuff themselves; then, when all three of them try to mix the same amount of ingredients at the same time the brew explodes.
|
{
"evidence": [
"Beer Barrel Polecats_sentence_2",
"Beer Barrel Polecats_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:\nUnable to find any bars selling beer, the Stooges opt to become bootleggers and brew some of the stuff themselves; then, when all three of them try to mix the same amount of ingredients at the same time the brew explodes.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Beer Barrel Polecats_sentence_2",
"Beer Barrel Polecats_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2744,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.057",
"value": "?search="
},
{
"operation": "search",
"time": "14.757",
"value": "Contains...Unable to find any bars selling beer, the Stooges opt to become bootlegger"
},
{
"operation": "Now on",
"time": "16.58",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "28.004",
"value": "Beer Barrel Polecats"
},
{
"operation": "Now on",
"time": "28.694",
"value": "Beer Barrel Polecats"
},
{
"operation": "Page search",
"time": "36.286",
"value": "Unable to find any bars selling"
},
{
"operation": "Highlighting",
"time": "38.134",
"value": "Beer Barrel Polecats_sentence_2"
},
{
"operation": "Highlighting",
"time": "47.196",
"value": "Beer Barrel Polecats_sentence_3"
},
{
"operation": "finish",
"time": "64.14",
"value": "finish"
}
]
}
|
35203
|
Pavel Belyayev was awarded the Distinguished Combat service medal (made of gold plated brass and lacquer) in 1953; then in 1954, he was promoted to captain.
|
{
"evidence": [
"Pavel Belyayev_sentence_26",
"Pavel Belyayev_sentence_27"
],
"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:\nPavel Belyayev was awarded the Distinguished Combat service medal (made of gold plated brass and lacquer) in 1953; then in 1954, he was promoted to captain.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Pavel Belyayev_sentence_26",
"Pavel Belyayev_sentence_27"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2745,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.118",
"value": "?search="
},
{
"operation": "search",
"time": "9.77",
"value": "Pavel Belyayev"
},
{
"operation": "Now on",
"time": "10.889",
"value": "Pavel Belyayev"
},
{
"operation": "hyperlink",
"time": "26.713",
"value": "6 Honors and awards Pavel Belyayev_section_6"
},
{
"operation": "Page search",
"time": "46.137",
"value": "Distinguished Combat"
},
{
"operation": "Highlighting",
"time": "50.944",
"value": "Pavel Belyayev_sentence_26"
},
{
"operation": "Highlighting",
"time": "59.437",
"value": "Pavel Belyayev_sentence_27"
},
{
"operation": "search",
"time": "73.045",
"value": "Distinguished Combat service medal"
},
{
"operation": "Now on",
"time": "75.831",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "86.159",
"value": "Air Force Distinguished Service Medal"
},
{
"operation": "Now on",
"time": "87.582",
"value": "Air Force Distinguished Service Medal"
},
{
"operation": "back-button-clicked",
"time": "113.317",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "114.838",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "120.752",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "120.757",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "120.889",
"value": "Pavel Belyayev#Honors and awards Pavel Belyayev section 6"
},
{
"operation": "Page search",
"time": "155.086",
"value": "made of"
},
{
"operation": "search",
"time": "170.733",
"value": "Soviet Air Forces"
},
{
"operation": "Now on",
"time": "172.956",
"value": "Soviet Air Forces"
},
{
"operation": "Page search",
"time": "180.074",
"value": "service medal"
},
{
"operation": "back-button-clicked",
"time": "191.035",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "191.184",
"value": "Pavel Belyayev#Honors and awards Pavel Belyayev section 6"
},
{
"operation": "finish",
"time": "199.988",
"value": "finish"
}
]
}
|
32371
|
On 7 April, Gneisenau carried the new ambassador to Japan, Arthur Graf Rex, from Taku (a city located in Saga Prefecture on the island of Kyushu) to Yokohama; then afterwards went on a tour of Japanese and Siberian waters.
|
{
"evidence": [
"SMS Gneisenau_sentence_60",
"SMS Gneisenau_sentence_61",
"Taku, Saga_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:\nOn 7 April, Gneisenau carried the new ambassador to Japan, Arthur Graf Rex, from Taku (a city located in Saga Prefecture on the island of Kyushu) to Yokohama; then afterwards went on a tour of Japanese and Siberian waters.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"SMS Gneisenau_sentence_60",
"SMS Gneisenau_sentence_61",
"Taku, Saga_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2746,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.39",
"value": "?search="
},
{
"operation": "search",
"time": "25.751",
"value": "Gneisenau"
},
{
"operation": "Now on",
"time": "26.79",
"value": "Gneisenau"
},
{
"operation": "hyperlink",
"time": "35.566",
"value": "August von Gneisenau"
},
{
"operation": "Now on",
"time": "36.669",
"value": "August Neidhardt von Gneisenau"
},
{
"operation": "back-button-clicked",
"time": "42.376",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "42.778",
"value": "Gneisenau"
},
{
"operation": "Highlighting",
"time": "44.369",
"value": "Gneisenau_item_0_1"
},
{
"operation": "Highlighting",
"time": "44.373",
"value": "Gneisenau_title"
},
{
"operation": "Highlighting deleted",
"time": "45.308",
"value": "Gneisenau_item_0_1"
},
{
"operation": "Highlighting deleted",
"time": "45.312",
"value": "Gneisenau_title"
},
{
"operation": "hyperlink",
"time": "56.29",
"value": "SS Gneisenau"
},
{
"operation": "Now on",
"time": "56.993",
"value": "SS Gneisenau (1935)"
},
{
"operation": "Page search",
"time": "69.669",
"value": "ambassador"
},
{
"operation": "back-button-clicked",
"time": "75.243",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "75.63",
"value": "Gneisenau"
},
{
"operation": "hyperlink",
"time": "86.979",
"value": "SMS Gneisenau"
},
{
"operation": "Now on",
"time": "87.931",
"value": "SMS Gneisenau (1879)"
},
{
"operation": "Page search",
"time": "95.317",
"value": "ambassador"
},
{
"operation": "Page search",
"time": "120.405",
"value": "Arthur Graf Rex"
},
{
"operation": "hyperlink",
"time": "140.195",
"value": "Gneisenau (disambiguation)"
},
{
"operation": "Now on",
"time": "141.213",
"value": "Gneisenau"
},
{
"operation": "Highlighting",
"time": "144.661",
"value": "Gneisenau_item_0_1"
},
{
"operation": "Highlighting",
"time": "144.664",
"value": "Gneisenau_title"
},
{
"operation": "Highlighting deleted",
"time": "147.252",
"value": "Gneisenau_item_0_1"
},
{
"operation": "Highlighting deleted",
"time": "147.258",
"value": "Gneisenau_title"
},
{
"operation": "hyperlink",
"time": "158.812",
"value": "SMS Gneisenau"
},
{
"operation": "Now on",
"time": "159.826",
"value": "SMS Gneisenau"
},
{
"operation": "Page search",
"time": "170.286",
"value": "taku"
},
{
"operation": "Highlighting",
"time": "176.673",
"value": "SMS Gneisenau_sentence_60"
},
{
"operation": "Highlighting",
"time": "198.513",
"value": "SMS Gneisenau_sentence_61"
},
{
"operation": "hyperlink",
"time": "209.574",
"value": "Taku"
},
{
"operation": "Now on",
"time": "210.685",
"value": "Taku, Saga"
},
{
"operation": "Highlighting",
"time": "212.776",
"value": "Taku, Saga_sentence_0"
},
{
"operation": "finish",
"time": "243.272",
"value": "finish"
}
]
}
|
22615
|
Alstermo, the birthplace of the Swedish actress Siw Carlsson, is located in the Uppvidinge Municipality of the Kronoberg County and has recorded a population of 829 in 2010.
|
{
"evidence": [
"Alstermo_sentence_0",
"Alstermo_sentence_13"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nAlstermo, the birthplace of the Swedish actress Siw Carlsson, is located in the Uppvidinge Municipality of the Kronoberg County and has recorded a population of 829 in 2010.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Alstermo_sentence_0",
"Alstermo_sentence_13"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2747,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.779",
"value": "?search="
},
{
"operation": "search",
"time": "8.413",
"value": "Alstermo"
},
{
"operation": "Now on",
"time": "9.231",
"value": "Alstermo"
},
{
"operation": "Highlighting",
"time": "28.206",
"value": "Alstermo_sentence_0"
},
{
"operation": "Highlighting",
"time": "38.29",
"value": "Alstermo_sentence_13"
},
{
"operation": "finish",
"time": "42.205",
"value": "finish"
}
]
}
|
54845
|
Jefrem was appointed in 1693 as the Orthodox bishop of Mohacs and Sziget in the Orthodox monastery of Branjina.
|
{
"evidence": [
"Jefrem Janković Tetovac_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:\nJefrem was appointed in 1693 as the Orthodox bishop of Mohacs and Sziget in the Orthodox monastery of Branjina.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Jefrem Janković Tetovac_sentence_7"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2748,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.139",
"value": "?search="
},
{
"operation": "search",
"time": "18.235",
"value": "Jefrem (patriarch)"
},
{
"operation": "Now on",
"time": "19.277",
"value": "Jefrem (patriarch)"
},
{
"operation": "search",
"time": "45.856",
"value": "Jefrem Janković Tetovac"
},
{
"operation": "Now on",
"time": "46.766",
"value": "Jefrem Janković Tetovac"
},
{
"operation": "Highlighting",
"time": "73.855",
"value": "Jefrem Janković Tetovac_sentence_7"
},
{
"operation": "finish",
"time": "82.721",
"value": "finish"
}
]
}
|
46312
|
There were only injuries reported in the August 21 accident of 1945 in Los Alamos.
|
{
"evidence": [
"Harry Daghlian_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:\nThere were only injuries reported in the August 21 accident of 1945 in Los Alamos.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Harry Daghlian_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2749,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.712",
"value": "?search="
},
{
"operation": "search",
"time": "28.396",
"value": "accident august 21 1945 los alamos"
},
{
"operation": "Now on",
"time": "30.401",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "43.461",
"value": "Los Alamos Manhattan Project_section_10"
},
{
"operation": "Now on",
"time": "44.763",
"value": "Manhattan Project#Los Alamos Manhattan Project section 10"
},
{
"operation": "search",
"time": "96.587",
"value": "los alamos accident 1945"
},
{
"operation": "Now on",
"time": "97.287",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "111.253",
"value": "Criticality accident Harry Daghlian_section_1"
},
{
"operation": "Now on",
"time": "111.805",
"value": "Harry Daghlian#Criticality accident Harry Daghlian section 1"
},
{
"operation": "Highlighting",
"time": "118.951",
"value": "Harry Daghlian_sentence_12"
},
{
"operation": "Highlighting",
"time": "118.953",
"value": "Harry Daghlian_section_1"
},
{
"operation": "Highlighting",
"time": "134.666",
"value": "Harry Daghlian_sentence_15"
},
{
"operation": "Highlighting",
"time": "138.374",
"value": "Harry Daghlian_sentence_17"
},
{
"operation": "Highlighting",
"time": "156.796",
"value": "Harry Daghlian_sentence_22"
},
{
"operation": "Highlighting",
"time": "190.547",
"value": "Harry Daghlian_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "200.195",
"value": "Harry Daghlian_sentence_22"
},
{
"operation": "Highlighting deleted",
"time": "201.21",
"value": "Harry Daghlian_sentence_17"
},
{
"operation": "Highlighting deleted",
"time": "202.055",
"value": "Harry Daghlian_sentence_15"
},
{
"operation": "Highlighting deleted",
"time": "203.844",
"value": "Harry Daghlian_sentence_12"
},
{
"operation": "Highlighting deleted",
"time": "203.846",
"value": "Harry Daghlian_section_1"
},
{
"operation": "finish",
"time": "222.178",
"value": "finish"
}
]
}
|
exfever_6054
|
Mr. Nobody is a 2009 science fiction drama film starring a Welsh actor and musician best known for his role in a 2004 British psychological thriller film.
|
{
"evidence": [
"Mr. Nobody (film)",
"Rhys Ifans",
"Enduring Love (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:\nMr. Nobody is a 2009 science fiction drama film starring a Welsh actor and musician best known for his role in a 2004 British psychological thriller film.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Mr. Nobody (film)",
"Rhys Ifans",
"Enduring Love (film)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2750,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
22727
|
Eilean Munde is a small uninhabited island in Loch Leven, Scotland and the site of a chapel built by St. Fintan Mundus with neighboring Eilean a' Chomhraidh(spelled Ylen na Ban) which is a meeting-place of those persons who had disputes
|
{
"evidence": [
"Eilean Munde_sentence_0",
"Eilean Munde_sentence_1",
"Eilean Munde_sentence_8",
"Eilean Munde_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:\nEilean Munde is a small uninhabited island in Loch Leven, Scotland and the site of a chapel built by St. Fintan Mundus with neighboring Eilean a' Chomhraidh(spelled Ylen na Ban) which is a meeting-place of those persons who had disputes\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Eilean Munde_sentence_0",
"Eilean Munde_sentence_1",
"Eilean Munde_sentence_8",
"Eilean Munde_sentence_9"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2751,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.305",
"value": "?search="
},
{
"operation": "search",
"time": "47.055",
"value": "Eilean Munde"
},
{
"operation": "Now on",
"time": "48.489",
"value": "Eilean Munde"
},
{
"operation": "Highlighting",
"time": "53.132",
"value": "Eilean Munde_sentence_0"
},
{
"operation": "Highlighting",
"time": "63.369",
"value": "Eilean Munde_sentence_1"
},
{
"operation": "Highlighting",
"time": "90.646",
"value": "Eilean Munde_sentence_8"
},
{
"operation": "Highlighting",
"time": "92.911",
"value": "Eilean Munde_sentence_9"
},
{
"operation": "finish",
"time": "123.855",
"value": "finish"
}
]
}
|
exfever_992
|
Jack Dylan Grazer is a Filipino actor who had a breakthrough playing the role of Eddie Kaspbrak in the film adaptation of the Stephen King novel It, directed by Andy Muschietti and written by Chase Palmer.
|
{
"evidence": [
"Jack Dylan Grazer",
"It (2017 film)"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJack Dylan Grazer is a Filipino actor who had a breakthrough playing the role of Eddie Kaspbrak in the film adaptation of the Stephen King novel It, directed by Andy Muschietti and written by Chase Palmer.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Jack Dylan Grazer",
"It (2017 film)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2752,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_216
|
The Dark Half was also nominated for Saturn Award for Best Horror Film.
|
{
"evidence": [
"The Dark Half (film)",
"Michael Rooker",
"Eight Men Out"
],
"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 Dark Half was also nominated for Saturn Award for Best Horror Film.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Dark Half (film)",
"Michael Rooker",
"Eight Men Out"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2753,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5962
|
Brendan passed away suddenly at home on 10th February 2022 due to a heart attack because of chain-smoking.
|
{
"evidence": [
"Neighbors (2014 film)",
"Dave Franco"
],
"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:\nBrendan passed away suddenly at home on 10th February 2022 due to a heart attack because of chain-smoking.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Neighbors (2014 film)",
"Dave Franco"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2754,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8177
|
The Han Chinese comprise approximately 95% of the population of Macao Special Administrative Region of the People's Republic of a country in East Asia.
|
{
"evidence": [
"Chinese people",
"Macau",
"China"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Han Chinese comprise approximately 95% of the population of Macao Special Administrative Region of the People's Republic of a country in East Asia.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Chinese people",
"Macau",
"China"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2755,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1409
|
Eva Ga?lle Green is a French actress who appeared in a 2006 epic historical fiction drama film that stars an English actor who began his acting career on stage in 1969 and has appeared in many West End theatre productions.
|
{
"evidence": [
"Eva Green",
"Kingdom of Heaven (film)",
"Jeremy Irons"
],
"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:\nEva Ga?lle Green is a French actress who appeared in a 2006 epic historical fiction drama film that stars an English actor who began his acting career on stage in 1969 and has appeared in many West End theatre productions.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Eva Green",
"Kingdom of Heaven (film)",
"Jeremy Irons"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2756,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
21652
|
The Charlotteville Cycling Club was formed at the Charlotteville Institute in Addison Road in 1903 and organizes the Guilford Town Centre Races.
|
{
"evidence": [
"Charlotteville Cycling Club_sentence_1",
"Charlotteville Cycling Club_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:\nThe Charlotteville Cycling Club was formed at the Charlotteville Institute in Addison Road in 1903 and organizes the Guilford Town Centre Races.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Charlotteville Cycling Club_sentence_1",
"Charlotteville Cycling Club_sentence_9"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2757,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.968",
"value": "?search="
},
{
"operation": "search",
"time": "14.643",
"value": "Charlotteville Cycling Club"
},
{
"operation": "Now on",
"time": "15.435",
"value": "Charlotteville Cycling Club"
},
{
"operation": "Highlighting",
"time": "24.956",
"value": "Charlotteville Cycling Club_sentence_1"
},
{
"operation": "Page search",
"time": "50.173",
"value": "addison"
},
{
"operation": "Highlighting",
"time": "57.703",
"value": "Charlotteville Cycling Club_sentence_9"
},
{
"operation": "finish",
"time": "68.964",
"value": "finish"
}
]
}
|
18290
|
The Vassar College Observatory not only served as an educational and research facility wherein students are usually found using both the main telescope and any of the smaller instruments, but also as a home for Maria and her father during and after her tenure as professor.
|
{
"evidence": [
"Vassar College Observatory_sentence_10",
"Vassar College Observatory_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:\nThe Vassar College Observatory not only served as an educational and research facility wherein students are usually found using both the main telescope and any of the smaller instruments, but also as a home for Maria and her father during and after her tenure as professor.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Vassar College Observatory_sentence_10",
"Vassar College Observatory_sentence_9"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2758,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.527",
"value": "?search="
},
{
"operation": "search",
"time": "17.298",
"value": "Vassar College Observatory"
},
{
"operation": "Now on",
"time": "18.708",
"value": "Vassar College Observatory"
},
{
"operation": "Page search",
"time": "50.38",
"value": "facility"
},
{
"operation": "Highlighting",
"time": "55.655",
"value": "Vassar College Observatory_sentence_10"
},
{
"operation": "Highlighting",
"time": "98.192",
"value": "Vassar College Observatory_sentence_9"
},
{
"operation": "finish",
"time": "118.328",
"value": "finish"
}
]
}
|
exfever_3711
|
Romelu Lukaku Bolingoli is a Canadian professional footballer who plays as a striker for the top level of the football pyramid, a series of interconnected leagues for men's association football clubs in England.
|
{
"evidence": [
"Romelu Lukaku",
"Premier League",
"English football league system"
],
"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:\nRomelu Lukaku Bolingoli is a Canadian professional footballer who plays as a striker for the top level of the football pyramid, a series of interconnected leagues for men's association football clubs in England.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Romelu Lukaku",
"Premier League",
"English football league system"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2759,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6903
|
Montreal is named after a large intrusive rock hill immediately west of Downtown Montreal, in one of the thirteen provinces and territories of Canada.
|
{
"evidence": [
"Montreal",
"Mount Royal",
"Quebec"
],
"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:\nMontreal is named after a large intrusive rock hill immediately west of Downtown Montreal, in one of the thirteen provinces and territories of Canada.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Montreal",
"Mount Royal",
"Quebec"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2760,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
83167
|
Vermont Route 29 which was originally called VT F-2 in 1926 is just 5.315 miles long, it starts in Isle La Motte and ends at Route 2 in Alburgh.
|
{
"evidence": [
"Vermont Route 129_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:\nVermont Route 29 which was originally called VT F-2 in 1926 is just 5.315 miles long, it starts in Isle La Motte and ends at Route 2 in Alburgh.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Vermont Route 129_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2761,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.852",
"value": "?search="
},
{
"operation": "search",
"time": "13.558",
"value": "Contains...Vermont Route 29"
},
{
"operation": "Now on",
"time": "15.256",
"value": "undefined"
},
{
"operation": "search",
"time": "75.331",
"value": "US ROUTE 29 IN VERMONT"
},
{
"operation": "Now on",
"time": "76.38",
"value": "undefined"
},
{
"operation": "search",
"time": "98.174",
"value": "STATE ROUTE 29 (VERMONT)"
},
{
"operation": "Now on",
"time": "99.508",
"value": "undefined"
},
{
"operation": "search",
"time": "126.634",
"value": "Contains...VT F-2"
},
{
"operation": "Now on",
"time": "129.446",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "135.252",
"value": "Vermont Route 129"
},
{
"operation": "Now on",
"time": "135.75",
"value": "Vermont Route 129"
},
{
"operation": "Highlighting",
"time": "159.513",
"value": "Vermont Route 129_sentence_0"
},
{
"operation": "finish",
"time": "188.297",
"value": "finish"
}
]
}
|
exfever_4832
|
Radiohead is a Polish rock band that worked with an English record producer, recording engineer, and musician.
|
{
"evidence": [
"Radiohead",
"Nigel Godrich"
],
"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:\nRadiohead is a Polish rock band that worked with an English record producer, recording engineer, and musician.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Radiohead",
"Nigel Godrich"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2762,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
66215
|
The Basketball at the 2019 Pan American Games has three events including the Olympic Champions and 2017 FIBA Women's AmeriCup.
|
{
"evidence": [
"Basketball at the 2019 Pan American Games_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 Basketball at the 2019 Pan American Games has three events including the Olympic Champions and 2017 FIBA Women's AmeriCup.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Basketball at the 2019 Pan American Games_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2763,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.068",
"value": "?search="
},
{
"operation": "search",
"time": "26.388",
"value": "2019 Pan American Games"
},
{
"operation": "Now on",
"time": "27.463",
"value": "2019 Pan American Games"
},
{
"operation": "search",
"time": "42.505",
"value": "Basketball at the 2019 Pan American Games"
},
{
"operation": "Now on",
"time": "43.52",
"value": "Basketball at the 2019 Pan American Games"
},
{
"operation": "Highlighting",
"time": "71.611",
"value": "Basketball at the 2019 Pan American Games_sentence_2"
},
{
"operation": "finish",
"time": "83.475",
"value": "finish"
}
]
}
|
29527
|
Warren Crandall Giles (May 28, 1896 – February 7, 1979) never played professional baseball nor was he ever inducted to the National Baseball Hall of Fame.
|
{
"evidence": [
"Warren Giles_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:\nWarren Crandall Giles (May 28, 1896 – February 7, 1979) never played professional baseball nor was he ever inducted to the National Baseball Hall of Fame.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Warren Giles_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2764,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.32",
"value": "?search="
},
{
"operation": "search",
"time": "15.562",
"value": "Warren Crandall Giles"
},
{
"operation": "Now on",
"time": "16.716",
"value": "Warren Giles"
},
{
"operation": "Highlighting",
"time": "28.637",
"value": "Warren Giles_sentence_1"
},
{
"operation": "finish",
"time": "48.238",
"value": "finish"
}
]
}
|
exfever_138
|
Amsterdam is found within the administrative layer between the national government and the local municipalities.
|
{
"evidence": [
"Amsterdam",
"Provinces of the Netherlands"
],
"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:\nAmsterdam is found within the administrative layer between the national government and the local municipalities.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Amsterdam",
"Provinces of the Netherlands"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2765,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7739
|
India is the seventh-largest country by area, it shares land borders to the north with the world's most populous country, with a population of more than 1.4 billion.
|
{
"evidence": [
"India",
"China"
],
"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:\nIndia is the seventh-largest country by area, it shares land borders to the north with the world's most populous country, with a population of more than 1.4 billion.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"India",
"China"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2766,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8490
|
Adrien Jerome Broner started his career in boxing at the age of 18.
|
{
"evidence": [
"Adrien Broner",
"Lightweight",
"Rowing (sport)"
],
"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:\nAdrien Jerome Broner started his career in boxing at the age of 18.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Adrien Broner",
"Lightweight",
"Rowing (sport)"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2767,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
39173
|
In Dunster Dovecote the priory was abolished in 1540 and was sold to the family from Dunster Castle, and in the 18th century the floor and door level was raised.
|
{
"evidence": [
"Dunster Dovecote_sentence_9"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nIn Dunster Dovecote the priory was abolished in 1540 and was sold to the family from Dunster Castle, and in the 18th century the floor and door level was raised.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Dunster Dovecote_sentence_9"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2768,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.679",
"value": "?search="
},
{
"operation": "search",
"time": "17.437",
"value": "Dunster Dovecote"
},
{
"operation": "Now on",
"time": "18.119",
"value": "Dunster Dovecote"
},
{
"operation": "Highlighting",
"time": "34.684",
"value": "Dunster Dovecote_sentence_9"
},
{
"operation": "finish",
"time": "47.336",
"value": "finish"
}
]
}
|
exfever_744
|
David Tennant is the youngest Scottish actor who rose to fame for his role in the BBC sci-fi series Doctor Who.
|
{
"evidence": [
"David Tennant",
"Giacomo Casanova"
],
"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:\nDavid Tennant is the youngest Scottish actor who rose to fame for his role in the BBC sci-fi series Doctor Who.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"David Tennant",
"Giacomo Casanova"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2769,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
49680
|
The 1999 Brandenburg state election was held to elect the 3rd Landtag of Bradenburg which is the unicameral legislature of the state of Bradenburg in Germany.
|
{
"evidence": [
"1999 Brandenburg state election_sentence_0",
"Landtag of Brandenburg_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 1999 Brandenburg state election was held to elect the 3rd Landtag of Bradenburg which is the unicameral legislature of the state of Bradenburg in Germany.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"1999 Brandenburg state election_sentence_0",
"Landtag of Brandenburg_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2770,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.977",
"value": "?search="
},
{
"operation": "search",
"time": "13.071",
"value": "1999 Brandenburg state election"
},
{
"operation": "Now on",
"time": "14.629",
"value": "1999 Brandenburg state election"
},
{
"operation": "Highlighting",
"time": "16.811",
"value": "1999 Brandenburg state election_sentence_0"
},
{
"operation": "hyperlink",
"time": "26.261",
"value": "Landtag of Brandenburg"
},
{
"operation": "Now on",
"time": "29.491",
"value": "Landtag of Brandenburg"
},
{
"operation": "Highlighting",
"time": "31.077",
"value": "Landtag of Brandenburg_sentence_0"
},
{
"operation": "finish",
"time": "52.354",
"value": "finish"
}
]
}
|
exfever_3973
|
"Wild Ones" is a song by Tramar Lacel Dillard, an American rapper that had a 2007 breakout single, "Low," which was number one for 12 weeks in the United States and broke the record for digital download sales at the time of its release.
|
{
"evidence": [
"Wild Ones (song)",
"Flo Rida"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\n\"Wild Ones\" is a song by Tramar Lacel Dillard, an American rapper that had a 2007 breakout single, \"Low,\" which was number one for 12 weeks in the United States and broke the record for digital download sales at the time of its release.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Wild Ones (song)",
"Flo Rida"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2771,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
21696
|
The law of the instrument is a cognitive bias which Lee Loevinger dubbed the law "Loevinger's law of irresistible use," in 1984 only after Warren Buffett criticized financial markets in 1967.
|
{
"evidence": [
"Law of the instrument_sentence_24",
"Law of the instrument_sentence_25"
],
"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 law of the instrument is a cognitive bias which Lee Loevinger dubbed the law \"Loevinger's law of irresistible use,\" in 1984 only after Warren Buffett criticized financial markets in 1967.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Law of the instrument_sentence_24",
"Law of the instrument_sentence_25"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2772,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.505",
"value": "?search="
},
{
"operation": "search",
"time": "21.781",
"value": "Law of the instrument"
},
{
"operation": "Now on",
"time": "23.107",
"value": "Law of the instrument"
},
{
"operation": "Highlighting",
"time": "33.178",
"value": "Law of the instrument_sentence_0"
},
{
"operation": "hyperlink",
"time": "68.742",
"value": "6 Lee Loevinger Law of the instrument_section_5"
},
{
"operation": "Highlighting",
"time": "79.961",
"value": "Law of the instrument_sentence_24"
},
{
"operation": "Highlighting",
"time": "95.313",
"value": "Law of the instrument_sentence_25"
},
{
"operation": "Highlighting deleted",
"time": "146.883",
"value": "Law of the instrument_sentence_25"
},
{
"operation": "Highlighting",
"time": "149.526",
"value": "Law of the instrument_sentence_25"
},
{
"operation": "Highlighting deleted",
"time": "153.116",
"value": "Law of the instrument_sentence_0"
},
{
"operation": "finish",
"time": "162.18",
"value": "finish"
}
]
}
|
exfever_6932
|
Peter Edward "Ginger" Baker was an English drummer, composer, songwriter, and co-founder and occasional vocalist of the rock band Cream, known as "rock's first superstar drummer" for pioneering the music genre that emerged in the late 1970s when musicians fused jazz harmony and improvised with rock, funk, and rhythm and blues.
|
{
"evidence": [
"Ginger Baker",
"Jazz fusion"
],
"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:\nPeter Edward \"Ginger\" Baker was an English drummer, composer, songwriter, and co-founder and occasional vocalist of the rock band Cream, known as \"rock's first superstar drummer\" for pioneering the music genre that emerged in the late 1970s when musicians fused jazz harmony and improvised with rock, funk, and rhythm and blues.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Ginger Baker",
"Jazz fusion"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2773,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3308
|
STS-96 was the first Space Shuttle mission flown by Space Shuttle Discovery.
|
{
"evidence": [
"STS-96",
"International Space Station"
],
"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:\nSTS-96 was the first Space Shuttle mission flown by Space Shuttle Discovery.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"STS-96",
"International Space Station"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2774,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5334
|
Lake Powell is an artificial reservoir on the Colorado River in Utah and the 6th largest and the 14th most populous of the 52 states.
|
{
"evidence": [
"Lake Powell",
"Arizona"
],
"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:\nLake Powell is an artificial reservoir on the Colorado River in Utah and the 6th largest and the 14th most populous of the 52 states.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Lake Powell",
"Arizona"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2775,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4340
|
The bonobo is an invulnerable great ape that is one of the two species making up the taxonomic rank Pan that is used in the biological classification of living and fossil organisms as well as viruses.
|
{
"evidence": [
"Bonobo",
"Genus"
],
"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 bonobo is an invulnerable great ape that is one of the two species making up the taxonomic rank Pan that is used in the biological classification of living and fossil organisms as well as viruses.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Bonobo",
"Genus"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2776,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
87172
|
Baijiu has been in production since dynastic times and is bottled for domestic consumption and for shipping internationally.
|
{
"evidence": [
"Baijiu_sentence_10"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBaijiu has been in production since dynastic times and is bottled for domestic consumption and for shipping internationally.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Baijiu_sentence_10"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2777,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.255",
"value": "?search="
},
{
"operation": "search",
"time": "9.706",
"value": "Baijiu"
},
{
"operation": "Now on",
"time": "11.42",
"value": "Baijiu"
},
{
"operation": "Page search",
"time": "24.48",
"value": "production"
},
{
"operation": "Page search",
"time": "35.869",
"value": "dynastic"
},
{
"operation": "Highlighting",
"time": "72.148",
"value": "Baijiu_sentence_10"
},
{
"operation": "Highlighting deleted",
"time": "90.057",
"value": "Baijiu_sentence_10"
},
{
"operation": "Highlighting",
"time": "98.001",
"value": "Baijiu_item_0_1"
},
{
"operation": "hyperlink",
"time": "118.443",
"value": "3.1 Global baijiu market Baijiu_section_6"
},
{
"operation": "Page search",
"time": "146.198",
"value": "domestic"
},
{
"operation": "Page search",
"time": "163.999",
"value": "bottled"
},
{
"operation": "Page search",
"time": "190.336",
"value": "shipping"
},
{
"operation": "Page search",
"time": "214.202",
"value": "international"
},
{
"operation": "Page search",
"time": "265.208",
"value": "production"
},
{
"operation": "Highlighting deleted",
"time": "313.424",
"value": "Baijiu_item_0_1"
},
{
"operation": "Highlighting",
"time": "321.643",
"value": "Baijiu_sentence_10"
},
{
"operation": "finish",
"time": "330.081",
"value": "finish"
}
]
}
|
exfever_2760
|
Terminator Genisys is a 2015 American science fiction action film written by an American screenwriter who wrote the script for the American neo-noir psychological horror film directed by Mel Gibson.
|
{
"evidence": [
"Terminator Genisys",
"Laeta Kalogridis",
"Shutter Island (film)"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nTerminator Genisys is a 2015 American science fiction action film written by an American screenwriter who wrote the script for the American neo-noir psychological horror film directed by Mel Gibson.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Terminator Genisys",
"Laeta Kalogridis",
"Shutter Island (film)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2778,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1198
|
J.Lo was the seventy-fifth album released by Epic Records in 2001.
|
{
"evidence": [
"J.Lo (album)",
"Troy Oliver"
],
"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:\nJ.Lo was the seventy-fifth album released by Epic Records in 2001.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"J.Lo (album)",
"Troy Oliver"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2779,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
46509
|
Zapatourism, a highly supported touristic phenomenon in Africa that has started to lose support, attracts tourists by the message, image, or policy of the Zapatista movement which claims to provide autonomy, dignity, and freedom of the local indigenous communities.
|
{
"evidence": [
"Zapaturismo_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:\nZapatourism, a highly supported touristic phenomenon in Africa that has started to lose support, attracts tourists by the message, image, or policy of the Zapatista movement which claims to provide autonomy, dignity, and freedom of the local indigenous communities.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Zapaturismo_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2780,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.339",
"value": "?search="
},
{
"operation": "search",
"time": "14.129",
"value": "Zapatourism"
},
{
"operation": "Now on",
"time": "17.827",
"value": "Zapatourism"
},
{
"operation": "Highlighting",
"time": "25.49",
"value": "Zapaturismo_sentence_0"
},
{
"operation": "finish",
"time": "55.185",
"value": "finish"
}
]
}
|
77310
|
Dohrighat is located in the Mau district which from 2011 had less than 10,000 people reported.
|
{
"evidence": [
"Dohrighat_sentence_21"
],
"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:\nDohrighat is located in the Mau district which from 2011 had less than 10,000 people reported.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Dohrighat_sentence_21"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2781,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.817",
"value": "?search="
},
{
"operation": "search",
"time": "20.401",
"value": "Dohrighat"
},
{
"operation": "Now on",
"time": "20.975",
"value": "Dohrighat"
},
{
"operation": "Highlighting",
"time": "115.733",
"value": "Dohrighat_sentence_0"
},
{
"operation": "Highlighting",
"time": "132.634",
"value": "Dohrighat_sentence_21"
},
{
"operation": "finish",
"time": "155.34",
"value": "finish"
}
]
}
|
exfever_5052
|
Sarah Michelle Gellar gained international acclaim for her role in the television network, which was referred to as the "Frog Network" after the former mascot, an animated cartoon character from Marvel Comics.
|
{
"evidence": [
"Sarah Michelle Gellar",
"The WB",
"Michigan J. Frog"
],
"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:\nSarah Michelle Gellar gained international acclaim for her role in the television network, which was referred to as the \"Frog Network\" after the former mascot, an animated cartoon character from Marvel Comics.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Sarah Michelle Gellar",
"The WB",
"Michigan J. Frog"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2782,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2592
|
Doctor Zhivago by Boris Pasternak is a relatively long work of related events or experiences, whether nonfictional or fictional, typically written in prose and published as a book.
|
{
"evidence": [
"Doctor Zhivago (novel)",
"Novel",
"Narrative"
],
"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:\nDoctor Zhivago by Boris Pasternak is a relatively long work of related events or experiences, whether nonfictional or fictional, typically written in prose and published as a book.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Doctor Zhivago (novel)",
"Novel",
"Narrative"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2783,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.