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 |
|---|---|---|---|---|---|---|---|---|
94244
|
Canadian curler, Bill Tetley died at the age of 69 after having a career that included being the skip of the 1975 Brier Championship Team; he did not play in the Senior Curling Championships later in life.
|
{
"evidence": [
"Bill Tetley_sentence_0",
"Bill Tetley_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:\nCanadian curler, Bill Tetley died at the age of 69 after having a career that included being the skip of the 1975 Brier Championship Team; he did not play in the Senior Curling Championships later in life.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Bill Tetley_sentence_0",
"Bill Tetley_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5684,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.331",
"value": "?search="
},
{
"operation": "search",
"time": "9.274",
"value": "Bill Tetley"
},
{
"operation": "Now on",
"time": "10.234",
"value": "Bill Tetley"
},
{
"operation": "Highlighting",
"time": "12.295",
"value": "Bill Tetley_sentence_0"
},
{
"operation": "Highlighting",
"time": "39.097",
"value": "Bill Tetley_sentence_1"
},
{
"operation": "Highlighting",
"time": "46.515",
"value": "Bill Tetley_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "52.657",
"value": "Bill Tetley_sentence_1"
},
{
"operation": "finish",
"time": "58.233",
"value": "finish"
}
]
}
|
87814
|
The film Red Cherry, which was critic voted as the Worst Movie of the year at the 1996 Golden Rooster Awards, introduced the characters named Luo Luo Xiaoman and Chuchu as two new students who had come to Russia.
|
{
"evidence": [
"Red Cherry_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 film Red Cherry, which was critic voted as the Worst Movie of the year at the 1996 Golden Rooster Awards, introduced the characters named Luo Luo Xiaoman and Chuchu as two new students who had come to Russia.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Red Cherry_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5685,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.691",
"value": "?search="
},
{
"operation": "search",
"time": "10.971",
"value": "Red Cherry"
},
{
"operation": "Now on",
"time": "11.813",
"value": "Red Cherry"
},
{
"operation": "Highlighting",
"time": "47.053",
"value": "Red Cherry_sentence_2"
},
{
"operation": "finish",
"time": "103",
"value": "finish"
}
]
}
|
exfever_6260
|
James Cromwell is an American actor and activist best known for the films including a 1997 American magical realism drama film written and directed by Frank Darabont and based on Stephen King's 1996 novel.
|
{
"evidence": [
"James Cromwell",
"The Green Mile (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:\nJames Cromwell is an American actor and activist best known for the films including a 1997 American magical realism drama film written and directed by Frank Darabont and based on Stephen King's 1996 novel.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"James Cromwell",
"The Green Mile (film)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5686,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5387
|
Yangon is the city that physically encompasses the government's offices and meeting places in the Yangon Region and is the largest city in Myanmar.
|
{
"evidence": [
"Yangon",
"Capital city"
],
"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:\nYangon is the city that physically encompasses the government's offices and meeting places in the Yangon Region and is the largest city in Myanmar.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Yangon",
"Capital city"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5687,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1817
|
Gone with the Wind was the best-selling novel in 1936.
|
{
"evidence": [
"Gone with the Wind (novel)",
"Reconstruction Era"
],
"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:\nGone with the Wind was the best-selling novel in 1936.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Gone with the Wind (novel)",
"Reconstruction Era"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5688,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
8584
|
Kent State has appeared in four bowl games but lost three times against other teams.
|
{
"evidence": [
"Kent State Golden Flashes football_sentence_113"
],
"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:\nKent State has appeared in four bowl games but lost three times against other teams.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Kent State Golden Flashes football_sentence_113"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5689,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.735",
"value": "?search="
},
{
"operation": "search",
"time": "5.237",
"value": "Kent State"
},
{
"operation": "Now on",
"time": "6.274",
"value": "Kent State"
},
{
"operation": "Page search",
"time": "15.142",
"value": "bowl"
},
{
"operation": "search",
"time": "112.606",
"value": "ken state university games"
},
{
"operation": "Now on",
"time": "114.619",
"value": "undefined"
},
{
"operation": "search",
"time": "135.339",
"value": "bowl games"
},
{
"operation": "Now on",
"time": "136.502",
"value": "Bowl game"
},
{
"operation": "hyperlink",
"time": "146.368",
"value": "List of NCAA football bowl records"
},
{
"operation": "Now on",
"time": "147.637",
"value": "List of NCAA Division I FBS football bowl records"
},
{
"operation": "hyperlink",
"time": "158.439",
"value": "Kent State"
},
{
"operation": "Now on",
"time": "159.524",
"value": "Kent State Golden Flashes football"
},
{
"operation": "Highlighting",
"time": "179.411",
"value": "Kent State Golden Flashes football_cell_3_1_0"
},
{
"operation": "Highlighting",
"time": "179.415",
"value": "Kent State Golden Flashes football_header_cell_3_0_0"
},
{
"operation": "Highlighting",
"time": "181.009",
"value": "Kent State Golden Flashes football_cell_3_1_2"
},
{
"operation": "Highlighting",
"time": "181.011",
"value": "Kent State Golden Flashes football_header_cell_3_0_2"
},
{
"operation": "Highlighting",
"time": "184.082",
"value": "Kent State Golden Flashes football_cell_3_1_4"
},
{
"operation": "Highlighting",
"time": "184.085",
"value": "Kent State Golden Flashes football_header_cell_3_0_4"
},
{
"operation": "Highlighting",
"time": "188.986",
"value": "Kent State Golden Flashes football_cell_3_2_0"
},
{
"operation": "Highlighting",
"time": "189.95",
"value": "Kent State Golden Flashes football_cell_3_2_2"
},
{
"operation": "Highlighting",
"time": "204.087",
"value": "Kent State Golden Flashes football_sentence_113"
},
{
"operation": "finish",
"time": "230.605",
"value": "finish"
}
]
}
|
exfever_8160
|
Glee is an American comedy-drama television series that is also classified as a film genre in which songs by the characters are interwoven into the narrative that is exemplified by the films of Busby Berkeley that were typified by lavish production numbers such as a pre-Code Warner Bros. musical film directed by Jason Brock with songs by Harry Warren.
|
{
"evidence": [
"Glee (TV series)",
"Musical film",
"Gold Diggers of 1933"
],
"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:\nGlee is an American comedy-drama television series that is also classified as a film genre in which songs by the characters are interwoven into the narrative that is exemplified by the films of Busby Berkeley that were typified by lavish production numbers such as a pre-Code Warner Bros. musical film directed by Jason Brock with songs by Harry Warren.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Glee (TV series)",
"Musical film",
"Gold Diggers of 1933"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5690,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3037
|
Starbucks is an American multinational chain of coffeehouses that serves different types of coffee, including a coffee-brewing method that originated in Italy.
|
{
"evidence": [
"Starbucks",
"Espresso"
],
"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:\nStarbucks is an American multinational chain of coffeehouses that serves different types of coffee, including a coffee-brewing method that originated in Italy.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Starbucks",
"Espresso"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5691,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8959
|
The bonobo is an endangered great ape found in a 500,000 km2 area of the Congo Basin the second-largest country of Central America.
|
{
"evidence": [
"Bonobo",
"Democratic Republic of the Congo"
],
"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 endangered great ape found in a 500,000 km2 area of the Congo Basin the second-largest country of Central America.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Bonobo",
"Democratic Republic of the Congo"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5692,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6666
|
Supernatural is an Indian dark fantasy drama television starring an American actor who rose to fame in the early 2000s after appearing on the television series Gilmore Girls.
|
{
"evidence": [
"Supernatural (U.S. TV series)",
"Jared Padalecki"
],
"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:\nSupernatural is an Indian dark fantasy drama television starring an American actor who rose to fame in the early 2000s after appearing on the television series Gilmore Girls.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Supernatural (U.S. TV series)",
"Jared Padalecki"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5693,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
23551
|
Saving Max is a novel written by American author Antoinette van Heugten, published by Mira Books on September 28, 2010, that is about a teenager with ODD syndrome who was accused of murdering another patient at a mental hospital.
|
{
"evidence": [
"Saving Max_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:\nSaving Max is a novel written by American author Antoinette van Heugten, published by Mira Books on September 28, 2010, that is about a teenager with ODD syndrome who was accused of murdering another patient at a mental hospital.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Saving Max_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5694,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.365",
"value": "?search="
},
{
"operation": "search",
"time": "10.465",
"value": "Saving Max"
},
{
"operation": "Now on",
"time": "11.176",
"value": "Saving Max"
},
{
"operation": "Highlighting",
"time": "81.2",
"value": "Saving Max_header_cell_0_4_0"
},
{
"operation": "Highlighting",
"time": "81.731",
"value": "Saving Max_cell_0_4_1"
},
{
"operation": "Highlighting",
"time": "81.745",
"value": "Saving Max_header_cell_0_4_0"
},
{
"operation": "Highlighting",
"time": "82.51",
"value": "Saving Max_header_cell_0_5_0"
},
{
"operation": "Highlighting",
"time": "83.209",
"value": "Saving Max_cell_0_5_1"
},
{
"operation": "Highlighting",
"time": "83.219",
"value": "Saving Max_header_cell_0_5_0"
},
{
"operation": "Highlighting",
"time": "135.165",
"value": "Saving Max_sentence_0"
},
{
"operation": "Highlighting",
"time": "149.072",
"value": "Saving Max_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "150.299",
"value": "Saving Max_sentence_0"
},
{
"operation": "Highlighting",
"time": "153.122",
"value": "Saving Max_sentence_0"
},
{
"operation": "hyperlink",
"time": "159.47",
"value": "Asperger syndrome"
},
{
"operation": "Now on",
"time": "160.385",
"value": "Asperger syndrome"
},
{
"operation": "Page search",
"time": "174.515",
"value": "odd"
},
{
"operation": "back-button-clicked",
"time": "184.192",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "184.278",
"value": "Saving Max"
},
{
"operation": "Highlighting deleted",
"time": "189.081",
"value": "Saving Max_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "190.588",
"value": "Saving Max_cell_0_4_1"
},
{
"operation": "Highlighting deleted",
"time": "190.6",
"value": "Saving Max_header_cell_0_4_0"
},
{
"operation": "Highlighting deleted",
"time": "191.349",
"value": "Saving Max_cell_0_5_1"
},
{
"operation": "Highlighting deleted",
"time": "191.357",
"value": "Saving Max_header_cell_0_5_0"
},
{
"operation": "finish",
"time": "293.674",
"value": "finish"
}
]
}
|
exfever_6118
|
Wilder also wrote his first script, adapting it from the Yves Robert film Pardon Mon Affaire.
|
{
"evidence": [
"The Woman in Red (1984 film)",
"Kelly LeBrock",
"Weird Science (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:\nWilder also wrote his first script, adapting it from the Yves Robert film Pardon Mon Affaire.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Woman in Red (1984 film)",
"Kelly LeBrock",
"Weird Science (film)"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5695,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1224
|
Priyanka Chopra was the fourth Indian actress who won?the Miss World 2000 pageant.
|
{
"evidence": [
"Priyanka Chopra",
"National Film Awards"
],
"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:\nPriyanka Chopra was the fourth Indian actress who won?the Miss World 2000 pageant.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Priyanka Chopra",
"National Film Awards"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5696,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5867
|
Pluto is a dwarf planet in the Kuiper belt defined by a governmental organization promoting astronomical research, outreach, education, and development through global cooperation.
|
{
"evidence": [
"Pluto",
"International Astronomical Union"
],
"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:\nPluto is a dwarf planet in the Kuiper belt defined by a governmental organization promoting astronomical research, outreach, education, and development through global cooperation.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Pluto",
"International Astronomical Union"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5697,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8973
|
Apocalypse Now is a 1979 American epic psychological war film starring an American actor who first became known for his role in a 1973 American neo-noir period crime drama film directed by Terrence Malick.
|
{
"evidence": [
"Apocalypse Now",
"Martin Sheen",
"Badlands (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:\nApocalypse Now is a 1979 American epic psychological war film starring an American actor who first became known for his role in a 1973 American neo-noir period crime drama film directed by Terrence Malick.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Apocalypse Now",
"Martin Sheen",
"Badlands (film)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5698,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8035
|
Ajab Prem Ki Ghazab Kahani is an Indian 2009 Hindi-language romantic comedy film and the unofficial adaptation of a 2005 Indian Telugu-language action film produced by D. Suresh.
|
{
"evidence": [
"Ajab Prem Ki Ghazab Kahani",
"Soggadu (2005 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:\nAjab Prem Ki Ghazab Kahani is an Indian 2009 Hindi-language romantic comedy film and the unofficial adaptation of a 2005 Indian Telugu-language action film produced by D. Suresh.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Ajab Prem Ki Ghazab Kahani",
"Soggadu (2005 film)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5699,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
12142
|
Cyclones that develop in the North Atlantic basin are monitored by the CPHC.
|
{
"evidence": [
"National Hurricane Center_sentence_9",
"Tropical cyclone basins_sentence_8"
],
"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:\nCyclones that develop in the North Atlantic basin are monitored by the CPHC.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"National Hurricane Center_sentence_9",
"Tropical cyclone basins_sentence_8"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5700,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.913",
"value": "?search="
},
{
"operation": "search",
"time": "10.438",
"value": "CPHC"
},
{
"operation": "Now on",
"time": "11.291",
"value": "undefined"
},
{
"operation": "search",
"time": "26.164",
"value": "cyclones in the North Atlantic basin"
},
{
"operation": "Now on",
"time": "27.495",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "32.619",
"value": "cyclone in the north atlantic basin"
},
{
"operation": "Now on",
"time": "33.472",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "44.866",
"value": "List of named storms (T)"
},
{
"operation": "Now on",
"time": "45.609",
"value": "List of named storms (T)"
},
{
"operation": "search",
"time": "70.642",
"value": "cyclones CPHC"
},
{
"operation": "Now on",
"time": "71.351",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "79.823",
"value": "List of Pacific hurricanes"
},
{
"operation": "Now on",
"time": "82.016",
"value": "List of Pacific hurricanes"
},
{
"operation": "Page search",
"time": "89.005",
"value": "CPHC"
},
{
"operation": "search",
"time": "126.856",
"value": "Central Pacific Hurricane Center"
},
{
"operation": "Now on",
"time": "128.297",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "138.95",
"value": "National Hurricane Center"
},
{
"operation": "Now on",
"time": "139.727",
"value": "National Hurricane Center"
},
{
"operation": "Highlighting",
"time": "162.974",
"value": "National Hurricane Center_sentence_9"
},
{
"operation": "Page search",
"time": "179.511",
"value": "CPHC"
},
{
"operation": "Highlighting",
"time": "202.315",
"value": "National Hurricane Center_sentence_56"
},
{
"operation": "Highlighting",
"time": "202.317",
"value": "National Hurricane Center_section_3"
},
{
"operation": "Highlighting",
"time": "202.318",
"value": "National Hurricane Center_section_4"
},
{
"operation": "Highlighting",
"time": "221.794",
"value": "National Hurricane Center_item_0_2"
},
{
"operation": "Highlighting",
"time": "221.796",
"value": "National Hurricane Center_section_10"
},
{
"operation": "Highlighting deleted",
"time": "222.449",
"value": "National Hurricane Center_item_0_2"
},
{
"operation": "Highlighting deleted",
"time": "222.453",
"value": "National Hurricane Center_section_10"
},
{
"operation": "back-button-clicked",
"time": "232.65",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "233.404",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "251.026",
"value": "Tropical cyclone warnings and watches"
},
{
"operation": "Now on",
"time": "252.215",
"value": "Tropical cyclone warnings and watches"
},
{
"operation": "Page search",
"time": "256.676",
"value": "CPHC"
},
{
"operation": "hyperlink",
"time": "334.229",
"value": "weather forecasting"
},
{
"operation": "Now on",
"time": "335.291",
"value": "Weather forecasting"
},
{
"operation": "search",
"time": "372.822",
"value": "cyclones that develop in the North Atlantic basin are monitored by the CPHC"
},
{
"operation": "Now on",
"time": "374.534",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "393.891",
"value": "Tropical cyclone basins"
},
{
"operation": "Now on",
"time": "394.753",
"value": "Tropical cyclone basins"
},
{
"operation": "hyperlink",
"time": "400.785",
"value": "2.1 North Atlantic Ocean Tropical cyclone basins_section_2"
},
{
"operation": "Highlighting",
"time": "413.426",
"value": "Tropical cyclone basins_sentence_8"
},
{
"operation": "hyperlink",
"time": "416.755",
"value": "National Hurricane Center"
},
{
"operation": "Now on",
"time": "417.207",
"value": "National Hurricane Center"
},
{
"operation": "finish",
"time": "513.746",
"value": "finish"
}
]
}
|
exfever_5395
|
Mount Everest is an 8,849-meter mountain located in the Himalaya Mountain Range.
|
{
"evidence": [
"K2",
"Pakistan"
],
"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:\nMount Everest is an 8,849-meter mountain located in the Himalaya Mountain Range.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"K2",
"Pakistan"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5701,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
28405
|
Chushul Chakzam, a suspension bridge which literally means “iron bridge”, was built in 1430 by Thang Tong Gyalpo—known as the “Iron Bridge Maker” who is a great Buddhist adept, architect, physician, and a few other professions.
|
{
"evidence": [
"Chushul Chakzam_sentence_0",
"Chushul Chakzam_sentence_1",
"Thang Tong Gyalpo_sentence_0",
"Thang Tong Gyalpo_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nChushul Chakzam, a suspension bridge which literally means “iron bridge”, was built in 1430 by Thang Tong Gyalpo—known as the “Iron Bridge Maker” who is a great Buddhist adept, architect, physician, and a few other professions.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Chushul Chakzam_sentence_0",
"Chushul Chakzam_sentence_1",
"Thang Tong Gyalpo_sentence_0",
"Thang Tong Gyalpo_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5702,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.082",
"value": "?search="
},
{
"operation": "search",
"time": "5.891",
"value": "Chushul Chakzam"
},
{
"operation": "Now on",
"time": "6.763",
"value": "Chushul Chakzam"
},
{
"operation": "Highlighting",
"time": "8.527",
"value": "Chushul Chakzam_sentence_0"
},
{
"operation": "Highlighting",
"time": "17.315",
"value": "Chushul Chakzam_sentence_1"
},
{
"operation": "hyperlink",
"time": "23.292",
"value": "Thang Tong Gyalpo"
},
{
"operation": "Now on",
"time": "24.039",
"value": "Thang Tong Gyalpo"
},
{
"operation": "Highlighting",
"time": "26.048",
"value": "Thang Tong Gyalpo_sentence_0"
},
{
"operation": "Highlighting",
"time": "39.971",
"value": "Thang Tong Gyalpo_sentence_2"
},
{
"operation": "back-button-clicked",
"time": "44.729",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "44.812",
"value": "Chushul Chakzam"
},
{
"operation": "finish",
"time": "49.456",
"value": "finish"
}
]
}
|
69584
|
Democrat candidate John Patrick Hopkins won the Chicago mayoral special election of 1893 with a narrow margin of victory against Republican George Bell Swift following the assassination of Mayor Carter Harrison Sr.
|
{
"evidence": [
"John Patrick Hopkins_sentence_14"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nDemocrat candidate John Patrick Hopkins won the Chicago mayoral special election of 1893 with a narrow margin of victory against Republican George Bell Swift following the assassination of Mayor Carter Harrison Sr.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"John Patrick Hopkins_sentence_14"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5703,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.169",
"value": "?search="
},
{
"operation": "Now on",
"time": "6.207",
"value": "?search="
},
{
"operation": "search",
"time": "13.305",
"value": "John Patrick Hopkins"
},
{
"operation": "Now on",
"time": "14.277",
"value": "John Patrick Hopkins"
},
{
"operation": "Highlighting",
"time": "58.422",
"value": "John Patrick Hopkins_sentence_14"
},
{
"operation": "hyperlink",
"time": "123.307",
"value": "Mayor of Chicago"
},
{
"operation": "Now on",
"time": "124.467",
"value": "Mayor of Chicago"
},
{
"operation": "Highlighting",
"time": "163.754",
"value": "Mayor of Chicago_cell_2_36_2"
},
{
"operation": "Highlighting",
"time": "163.757",
"value": "Mayor of Chicago_title"
},
{
"operation": "Highlighting",
"time": "163.762",
"value": "Mayor of Chicago_section_3"
},
{
"operation": "Highlighting",
"time": "163.765",
"value": "Mayor of Chicago_section_5"
},
{
"operation": "Highlighting",
"time": "163.789",
"value": "Mayor of Chicago_header_cell_2_0_2"
},
{
"operation": "Highlighting deleted",
"time": "193.906",
"value": "Mayor of Chicago_cell_2_36_2"
},
{
"operation": "Highlighting deleted",
"time": "193.913",
"value": "Mayor of Chicago_title"
},
{
"operation": "Highlighting deleted",
"time": "193.919",
"value": "Mayor of Chicago_section_3"
},
{
"operation": "Highlighting deleted",
"time": "193.926",
"value": "Mayor of Chicago_section_5"
},
{
"operation": "Highlighting deleted",
"time": "193.951",
"value": "Mayor of Chicago_header_cell_2_0_2"
},
{
"operation": "hyperlink",
"time": "196.576",
"value": "George Bell Swift"
},
{
"operation": "Now on",
"time": "198.548",
"value": "George Bell Swift"
},
{
"operation": "finish",
"time": "245.287",
"value": "finish"
}
]
}
|
26046
|
Woodruff, a quarterback and punter at Kent State University, joined the University of Washington staff in Seattle as a graduate assistant under head coach Don James in 1980 and was also the head coach at Eastern Michigan University.
|
{
"evidence": [
"Jeff Woodruff_sentence_2",
"Jeff Woodruff_sentence_3",
"Jeff Woodruff_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:\nWoodruff, a quarterback and punter at Kent State University, joined the University of Washington staff in Seattle as a graduate assistant under head coach Don James in 1980 and was also the head coach at Eastern Michigan University.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Jeff Woodruff_sentence_2",
"Jeff Woodruff_sentence_3",
"Jeff Woodruff_sentence_7"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5704,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.679",
"value": "?search="
},
{
"operation": "search",
"time": "20.12",
"value": "Jeff Woodruff"
},
{
"operation": "Now on",
"time": "20.825",
"value": "Jeff Woodruff"
},
{
"operation": "Highlighting",
"time": "24.069",
"value": "Jeff Woodruff_header_cell_0_9_0"
},
{
"operation": "Highlighting",
"time": "25.055",
"value": "Jeff Woodruff_cell_0_9_1"
},
{
"operation": "Highlighting",
"time": "25.062",
"value": "Jeff Woodruff_header_cell_0_9_0"
},
{
"operation": "Highlighting",
"time": "25.063",
"value": "Jeff Woodruff_header_cell_0_6_0"
},
{
"operation": "Highlighting",
"time": "27.841",
"value": "Jeff Woodruff_header_cell_0_7_0"
},
{
"operation": "Highlighting",
"time": "28.317",
"value": "Jeff Woodruff_cell_0_7_1"
},
{
"operation": "Highlighting",
"time": "28.324",
"value": "Jeff Woodruff_header_cell_0_7_0"
},
{
"operation": "Highlighting deleted",
"time": "34.433",
"value": "Jeff Woodruff_cell_0_7_1"
},
{
"operation": "Highlighting deleted",
"time": "34.44",
"value": "Jeff Woodruff_header_cell_0_7_0"
},
{
"operation": "Highlighting",
"time": "36.965",
"value": "Jeff Woodruff_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "38.067",
"value": "Jeff Woodruff_cell_0_9_1"
},
{
"operation": "Highlighting deleted",
"time": "38.074",
"value": "Jeff Woodruff_header_cell_0_9_0"
},
{
"operation": "Highlighting deleted",
"time": "38.075",
"value": "Jeff Woodruff_header_cell_0_6_0"
},
{
"operation": "Highlighting",
"time": "42.141",
"value": "Jeff Woodruff_sentence_3"
},
{
"operation": "Page search",
"time": "71.638",
"value": "Michigan"
},
{
"operation": "Page search",
"time": "71.853",
"value": "Michigan"
},
{
"operation": "Highlighting",
"time": "74.161",
"value": "Jeff Woodruff_sentence_7"
},
{
"operation": "finish",
"time": "83.224",
"value": "finish"
}
]
}
|
exfever_965
|
The Getaway is the eleventh album by the band Red Hot Chili Peppers, whose genre includes popular music that usually uses instruments like the lowest-pitched member of the drum family.
|
{
"evidence": [
"The Getaway (Red Hot Chili Peppers album)",
"Rock music",
"Bass guitar"
],
"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 Getaway is the eleventh album by the band Red Hot Chili Peppers, whose genre includes popular music that usually uses instruments like the lowest-pitched member of the drum family.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Getaway (Red Hot Chili Peppers album)",
"Rock music",
"Bass guitar"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5705,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8698
|
Chet Atkins was an American dancer whose picking style are influenced by some of well-known guitarist including an American jazz, country, and blues guitarist that became the pioneer of the solid-body electric guitar, and his prototype, called the Log.
|
{
"evidence": [
"Chet Atkins",
"Les Paul"
],
"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:\nChet Atkins was an American dancer whose picking style are influenced by some of well-known guitarist including an American jazz, country, and blues guitarist that became the pioneer of the solid-body electric guitar, and his prototype, called the Log.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Chet Atkins",
"Les Paul"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5706,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5198
|
The House of Lusignan was the most renowned royal house in the world.
|
{
"evidence": [
"House of Lusignan",
"Kingdom of England",
"Great Britain"
],
"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 House of Lusignan was the most renowned royal house in the world.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"House of Lusignan",
"Kingdom of England",
"Great Britain"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5707,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
62898
|
Ilan Shohat (born 21 September 1974), was Israel's youngest politician to hold public office in 2008.
|
{
"evidence": [
"Ilan Shohat_sentence_1",
"Ilan Shohat_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:\nIlan Shohat (born 21 September 1974), was Israel's youngest politician to hold public office in 2008.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Ilan Shohat_sentence_1",
"Ilan Shohat_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5708,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.971",
"value": "?search="
},
{
"operation": "search",
"time": "37.113",
"value": "Ilan Shohat"
},
{
"operation": "Now on",
"time": "37.7",
"value": "Ilan Shohat"
},
{
"operation": "Highlighting",
"time": "47.098",
"value": "Ilan Shohat_sentence_1"
},
{
"operation": "Highlighting",
"time": "52.064",
"value": "Ilan Shohat_sentence_0"
},
{
"operation": "finish",
"time": "158.327",
"value": "finish"
}
]
}
|
exfever_4990
|
The Twelfth Doctor is an incarnation of the Doctor, which is a millennia-old alien from an ancient race of extraterrestrial people in the British science fiction television series Doctor Who.
|
{
"evidence": [
"Twelfth Doctor",
"Time Lord"
],
"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 Twelfth Doctor is an incarnation of the Doctor, which is a millennia-old alien from an ancient race of extraterrestrial people in the British science fiction television series Doctor Who.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Twelfth Doctor",
"Time Lord"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5709,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3353
|
Kathleen Erin Walsh is best known for her role as Dr. Addison Montgomery on the television dramas Grey's Anatomy aired by an American multinational commercial broadcast television network under an American multinational entertainment and media conglomerate.
|
{
"evidence": [
"Kate Walsh (actress)",
"American Broadcasting Company",
"The Walt Disney Company"
],
"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:\nKathleen Erin Walsh is best known for her role as Dr. Addison Montgomery on the television dramas Grey's Anatomy aired by an American multinational commercial broadcast television network under an American multinational entertainment and media conglomerate.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Kate Walsh (actress)",
"American Broadcasting Company",
"The Walt Disney Company"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5710,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
24433
|
In the nucleus accumbens, JunD directly opposes many of the neurological changes that occur in addiction, but has not been observed in humans.
|
{
"evidence": [
"JunD_sentence_6",
"JunD_sentence_8"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nIn the nucleus accumbens, JunD directly opposes many of the neurological changes that occur in addiction, but has not been observed in humans.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"JunD_sentence_6",
"JunD_sentence_8"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5711,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.027",
"value": "?search="
},
{
"operation": "search",
"time": "7.468",
"value": "JunD"
},
{
"operation": "Now on",
"time": "8.133",
"value": "JunD"
},
{
"operation": "search",
"time": "44.341",
"value": "Nucleus accumbens"
},
{
"operation": "Now on",
"time": "45.131",
"value": "Nucleus accumbens"
},
{
"operation": "Page search",
"time": "49.126",
"value": "JunD"
},
{
"operation": "back-button-clicked",
"time": "86.25",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "86.313",
"value": "JunD"
},
{
"operation": "Page search",
"time": "93.363",
"value": "oppose"
},
{
"operation": "Highlighting",
"time": "97.506",
"value": "JunD_sentence_6"
},
{
"operation": "Highlighting",
"time": "116.275",
"value": "JunD_sentence_8"
},
{
"operation": "finish",
"time": "176.847",
"value": "finish"
}
]
}
|
exfever_8417
|
David LeRoy Anderson is the second American makeup artist who received Oscar nominations.
|
{
"evidence": [
"David LeRoy Anderson",
"Rick Baker"
],
"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 LeRoy Anderson is the second American makeup artist who received Oscar nominations.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"David LeRoy Anderson",
"Rick Baker"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5712,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3102
|
To Die For is a 1995 film based on a song of the same name by an American novelist and journalist born on November 5, 1953.
|
{
"evidence": [
"To Die For",
"Joyce Maynard"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nTo Die For is a 1995 film based on a song of the same name by an American novelist and journalist born on November 5, 1953.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"To Die For",
"Joyce Maynard"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5713,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
57301
|
Monreith Estate is covered about 16,000 acres (65 km), and the Monreith House looks towards the White Loch of Myrton.
|
{
"evidence": [
"Glasserton_sentence_22",
"Glasserton_sentence_23"
],
"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:\nMonreith Estate is covered about 16,000 acres (65 km), and the Monreith House looks towards the White Loch of Myrton.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Glasserton_sentence_22",
"Glasserton_sentence_23"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5714,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.156",
"value": "?search="
},
{
"operation": "search",
"time": "491.296",
"value": "Glasserton"
},
{
"operation": "Now on",
"time": "492.251",
"value": "Glasserton"
},
{
"operation": "Highlighting",
"time": "534.951",
"value": "Glasserton_sentence_22"
},
{
"operation": "Highlighting",
"time": "560.661",
"value": "Glasserton_sentence_23"
},
{
"operation": "finish",
"time": "596.725",
"value": "finish"
}
]
}
|
exfever_6968
|
The Vampire Diaries is an American drama television series developed by a Canadian screenwriter, who is also known for creating the teen drama series Dawson's Creek for an American television network, first launched on broadcast television on January 11, 1995.
|
{
"evidence": [
"The Vampire Diaries",
"Kevin Williamson (screenwriter)",
"The WB"
],
"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 Vampire Diaries is an American drama television series developed by a Canadian screenwriter, who is also known for creating the teen drama series Dawson's Creek for an American television network, first launched on broadcast television on January 11, 1995.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Vampire Diaries",
"Kevin Williamson (screenwriter)",
"The WB"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5715,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5564
|
The House of the Spirits is a 1993 period drama film starring a British actor born on August 10, 1960.
|
{
"evidence": [
"The House of the Spirits (film)",
"Antonio Banderas"
],
"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 House of the Spirits is a 1993 period drama film starring a British actor born on August 10, 1960.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The House of the Spirits (film)",
"Antonio Banderas"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5716,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
47608
|
Vangjush Mio (an Albanian film director) was born 3 March 1891 and was known for painting, his notable work was Vjeshtë në Drenovë, Tirana
|
{
"evidence": [
"Vangjush Mio_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:\nVangjush Mio (an Albanian film director) was born 3 March 1891 and was known for painting, his notable work was Vjeshtë në Drenovë, Tirana\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Vangjush Mio_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5717,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.705",
"value": "?search="
},
{
"operation": "search",
"time": "35.653",
"value": "Vangjush Mio"
},
{
"operation": "Now on",
"time": "36.43",
"value": "Vangjush Mio"
},
{
"operation": "Highlighting",
"time": "48.406",
"value": "Vangjush Mio_sentence_0"
},
{
"operation": "finish",
"time": "154.127",
"value": "finish"
}
]
}
|
81294
|
The South Florida rocklands eco region covers 2,100 kilomters and is located in Florida and the Florida Keys, where their forests form on limestone outcrops.
|
{
"evidence": [
"South Florida rocklands_sentence_0",
"South Florida rocklands_sentence_1"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe South Florida rocklands eco region covers 2,100 kilomters and is located in Florida and the Florida Keys, where their forests form on limestone outcrops.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"South Florida rocklands_sentence_0",
"South Florida rocklands_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5718,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.329",
"value": "?search="
},
{
"operation": "search",
"time": "13.238",
"value": "South Florida rocklands"
},
{
"operation": "Now on",
"time": "14.321",
"value": "South Florida rocklands"
},
{
"operation": "Highlighting",
"time": "24.283",
"value": "South Florida rocklands_sentence_0"
},
{
"operation": "Highlighting",
"time": "34.875",
"value": "South Florida rocklands_sentence_1"
},
{
"operation": "finish",
"time": "44.16",
"value": "finish"
}
]
}
|
exfever_3835
|
Jared Padalecki is an African actor who is best known for playing the role of a fictional character and one of the protagonists in the American drama television series Supernatural along with his older brother, Dean.
|
{
"evidence": [
"Jared Padalecki",
"Sam Winchester"
],
"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:\nJared Padalecki is an African actor who is best known for playing the role of a fictional character and one of the protagonists in the American drama television series Supernatural along with his older brother, Dean.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Jared Padalecki",
"Sam Winchester"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5719,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7631
|
Shenzhen is a city and special administrative region of China in the eastern region of one of the most densely urbanized regions in the world and is often considered a megacity in South China.
|
{
"evidence": [
"Hong Kong",
"Pearl River Delta"
],
"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:\nShenzhen is a city and special administrative region of China in the eastern region of one of the most densely urbanized regions in the world and is often considered a megacity in South China.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Hong Kong",
"Pearl River Delta"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5720,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
49704
|
Robinson Crusoe Island (native name Isla Róbinson Crusoe) is part of the Juan Fernández Islands, a sparsely inhabited island group in the South Pacific Ocean reliant on tourism and fishing.
|
{
"evidence": [
"Robinson Crusoe Island_sentence_3",
"Robinson Crusoe Island_sentence_4",
"Robinson Crusoe Island_sentence_8",
"Robinson Crusoe Island_sentence_57"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nRobinson Crusoe Island (native name Isla Róbinson Crusoe) is part of the Juan Fernández Islands, a sparsely inhabited island group in the South Pacific Ocean reliant on tourism and fishing.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Robinson Crusoe Island_sentence_3",
"Robinson Crusoe Island_sentence_4",
"Robinson Crusoe Island_sentence_8",
"Robinson Crusoe Island_sentence_57"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5721,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.744",
"value": "?search="
},
{
"operation": "search",
"time": "48.298",
"value": "Robinson Crusoe Island"
},
{
"operation": "Now on",
"time": "49.292",
"value": "Robinson Crusoe Island"
},
{
"operation": "Highlighting",
"time": "99.179",
"value": "Robinson Crusoe Island_sentence_3"
},
{
"operation": "Highlighting",
"time": "100.655",
"value": "Robinson Crusoe Island_sentence_4"
},
{
"operation": "Highlighting",
"time": "103.409",
"value": "Robinson Crusoe Island_header_cell_0_4_0"
},
{
"operation": "Highlighting",
"time": "103.872",
"value": "Robinson Crusoe Island_cell_0_4_1"
},
{
"operation": "Highlighting",
"time": "103.889",
"value": "Robinson Crusoe Island_header_cell_0_4_0"
},
{
"operation": "Highlighting",
"time": "103.89",
"value": "Robinson Crusoe Island_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "103.89",
"value": "Robinson Crusoe Island_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "116.161",
"value": "Robinson Crusoe Island_cell_0_4_1"
},
{
"operation": "Highlighting deleted",
"time": "116.171",
"value": "Robinson Crusoe Island_header_cell_0_4_0"
},
{
"operation": "Highlighting deleted",
"time": "116.173",
"value": "Robinson Crusoe Island_header_cell_0_1_0"
},
{
"operation": "Highlighting deleted",
"time": "116.175",
"value": "Robinson Crusoe Island_header_cell_0_0_0"
},
{
"operation": "Page search",
"time": "213.766",
"value": "fishing"
},
{
"operation": "Page search",
"time": "307.293",
"value": "fish"
},
{
"operation": "Highlighting",
"time": "314.372",
"value": "Robinson Crusoe Island_sentence_8"
},
{
"operation": "Highlighting",
"time": "347.613",
"value": "Robinson Crusoe Island_sentence_57"
},
{
"operation": "hyperlink",
"time": "376.495",
"value": "Juan Fernández Islands"
},
{
"operation": "Now on",
"time": "377.726",
"value": "Juan Fernández Islands"
},
{
"operation": "back-button-clicked",
"time": "410.222",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "410.343",
"value": "Robinson Crusoe Island"
},
{
"operation": "finish",
"time": "433.027",
"value": "finish"
}
]
}
|
76790
|
On 7 April, Gneisenau carried the new ambassador to Japan, Arthur Graf Rex, from Taku to Yokohama; then afterwards went on a tour of Japanese and Siberian waters.
|
{
"evidence": [
"SMS Gneisenau_sentence_60",
"SMS Gneisenau_sentence_61"
],
"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 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"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5722,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.741",
"value": "?search="
},
{
"operation": "search",
"time": "28.297",
"value": "Gneisenau"
},
{
"operation": "Now on",
"time": "28.968",
"value": "Gneisenau"
},
{
"operation": "hyperlink",
"time": "60.299",
"value": "SS Gneisenau"
},
{
"operation": "Now on",
"time": "60.799",
"value": "SS Gneisenau (1935)"
},
{
"operation": "back-button-clicked",
"time": "76.804",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "76.854",
"value": "Gneisenau"
},
{
"operation": "hyperlink",
"time": "80.639",
"value": "SMS Gneisenau"
},
{
"operation": "Now on",
"time": "81.645",
"value": "SMS Gneisenau (1879)"
},
{
"operation": "back-button-clicked",
"time": "89.415",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "89.464",
"value": "Gneisenau"
},
{
"operation": "hyperlink",
"time": "94.027",
"value": "SMS Gneisenau"
},
{
"operation": "Now on",
"time": "95.144",
"value": "SMS Gneisenau"
},
{
"operation": "Highlighting",
"time": "98.921",
"value": "SMS Gneisenau_sentence_60"
},
{
"operation": "Highlighting",
"time": "105.873",
"value": "SMS Gneisenau_sentence_61"
},
{
"operation": "finish",
"time": "119.576",
"value": "finish"
}
]
}
|
exfever_7649
|
Evolution was the most sold album by Sabrina Carpenter.
|
{
"evidence": [
"Evolution (Sabrina Carpenter album)",
"Lead single"
],
"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:\nEvolution was the most sold album by Sabrina Carpenter.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Evolution (Sabrina Carpenter album)",
"Lead single"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5723,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
85274
|
Sulfite reductases do not belong to the oxidoreductase family, and participates in selenoamino acid metabolism and sulfur assimilation.
|
{
"evidence": [
"Sulfite reductase_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:\nSulfite reductases do not belong to the oxidoreductase family, and participates in selenoamino acid metabolism and sulfur assimilation.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Sulfite reductase_sentence_3"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5724,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.782",
"value": "?search="
},
{
"operation": "search",
"time": "10.73",
"value": "sulfite reductases"
},
{
"operation": "Now on",
"time": "11.757",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "57.341",
"value": "Sulfite reductase"
},
{
"operation": "Now on",
"time": "58.098",
"value": "Sulfite reductase"
},
{
"operation": "Highlighting",
"time": "83.24",
"value": "Sulfite reductase_sentence_3"
},
{
"operation": "finish",
"time": "91.014",
"value": "finish"
}
]
}
|
exfever_2706
|
Tobias Vincent Maguire is an American actor who has done films, including the film, set during the American Civil War, which follows a group of men who join the First Irregulars of the state in the Midwestern region of the United Kingdom.
|
{
"evidence": [
"Tobey Maguire",
"Ride with the Devil (film)",
"Missouri"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nTobias Vincent Maguire is an American actor who has done films, including the film, set during the American Civil War, which follows a group of men who join the First Irregulars of the state in the Midwestern region of the United Kingdom.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tobey Maguire",
"Ride with the Devil (film)",
"Missouri"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5725,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7700
|
Zootopia is a film produced by a Japanese animation studio that creates animated short films and is headquartered in a city in the southeastern end of the San Fernando Valley in Los Angeles County, California.
|
{
"evidence": [
"Zootopia",
"Walt Disney Animation Studios",
"Burbank, California"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nZootopia is a film produced by a Japanese animation studio that creates animated short films and is headquartered in a city in the southeastern end of the San Fernando Valley in Los Angeles County, California.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Zootopia",
"Walt Disney Animation Studios",
"Burbank, California"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5726,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4299
|
The Georgia Institute of Technology has a satellite campus in the oldest city in the U.S. state and was a strategic port city during the ideological and political revolution that occurred in British America between 1765 and 1791.
|
{
"evidence": [
"Georgia Institute of Technology",
"Savannah, Georgia",
"American Revolution"
],
"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 Georgia Institute of Technology has a satellite campus in the oldest city in the U.S. state and was a strategic port city during the ideological and political revolution that occurred in British America between 1765 and 1791.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Georgia Institute of Technology",
"Savannah, Georgia",
"American Revolution"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5727,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2680
|
Stranger than Fiction is a 2006 film directed by Marc Forster and starring a European actor who established himself in the mid-1990s as a cast member on the NBC sketch comedy show Saturday Night Live.
|
{
"evidence": [
"Stranger than Fiction (2006 film)",
"Will Ferrell"
],
"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:\nStranger than Fiction is a 2006 film directed by Marc Forster and starring a European actor who established himself in the mid-1990s as a cast member on the NBC sketch comedy show Saturday Night Live.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Stranger than Fiction (2006 film)",
"Will Ferrell"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5728,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7873
|
The screenplay of The Hunger Games film in 2012 was co-written by an American screenwriter and film director with numerous films to his credit.
|
{
"evidence": [
"The Hunger Games (film)",
"Billy Ray (screenwriter)"
],
"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 screenplay of The Hunger Games film in 2012 was co-written by an American screenwriter and film director with numerous films to his credit.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Hunger Games (film)",
"Billy Ray (screenwriter)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5729,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3358
|
Tropic Thunder is a 2008 action-comedy film starring a Canadian actor, director, stand-up comedian, and screenwriter known for the 2010 American action-fantasy film produced by Jerry Bruckheimer, directed by Jon Turteltaub, and released by Walt Disney Pictures,
|
{
"evidence": [
"Tropic Thunder",
"Jay Baruchel",
"The Sorcerer's Apprentice (2010 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:\nTropic Thunder is a 2008 action-comedy film starring a Canadian actor, director, stand-up comedian, and screenwriter known for the 2010 American action-fantasy film produced by Jerry Bruckheimer, directed by Jon Turteltaub, and released by Walt Disney Pictures,\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tropic Thunder",
"Jay Baruchel",
"The Sorcerer's Apprentice (2010 film)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5730,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3445
|
Henry Cavill is known for his portrayal in Showtime's The Tudors as the 1st Duke of Suffolk, an English military leader and courtier.
|
{
"evidence": [
"Henry Cavill",
"Charles Brandon, 1st Duke of Suffolk"
],
"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 Cavill is known for his portrayal in Showtime's The Tudors as the 1st Duke of Suffolk, an English military leader and courtier.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Henry Cavill",
"Charles Brandon, 1st Duke of Suffolk"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5731,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
90057
|
Betty Mars, a pop singer, was born in Marseilles, France's capital and its most populated city.
|
{
"evidence": [
"Paris_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:\nBetty Mars, a pop singer, was born in Marseilles, France's capital and its most populated city.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Paris_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5732,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.534",
"value": "?search="
},
{
"operation": "search",
"time": "22.175",
"value": "paris"
},
{
"operation": "Now on",
"time": "24.967",
"value": "Paris"
},
{
"operation": "Highlighting",
"time": "30.362",
"value": "Paris_sentence_2"
},
{
"operation": "finish",
"time": "40.918",
"value": "finish"
}
]
}
|
exfever_6464
|
The Falling is a 2014 British mystery drama film starring Florence Pugh and an English actress who led a character in the HBO epic medieval fantasy drama series Game of Thrones (2011–2019).
|
{
"evidence": [
"The Falling (2014 film)",
"Maisie Williams"
],
"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 Falling is a 2014 British mystery drama film starring Florence Pugh and an English actress who led a character in the HBO epic medieval fantasy drama series Game of Thrones (2011–2019).\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Falling (2014 film)",
"Maisie Williams"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5733,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
77558
|
Gabriel René Paul graduated in 1834 from the United States Military Academy, a four-year federal service academy in West Point, New York, and served as an officer in the 7th U.S. Infantry during the Seminole Wars and the Mexican–American War.
|
{
"evidence": [
"Gabriel René Paul_sentence_3",
"Gabriel René Paul_sentence_4",
"United States Military Academy_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nGabriel René Paul graduated in 1834 from the United States Military Academy, a four-year federal service academy in West Point, New York, and served as an officer in the 7th U.S. Infantry during the Seminole Wars and the Mexican–American War.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Gabriel René Paul_sentence_3",
"Gabriel René Paul_sentence_4",
"United States Military Academy_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5734,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.3",
"value": "?search="
},
{
"operation": "search",
"time": "14.893",
"value": "Gabriel René Paul"
},
{
"operation": "Now on",
"time": "15.819",
"value": "Gabriel René Paul"
},
{
"operation": "Highlighting",
"time": "43.54",
"value": "Gabriel René Paul_sentence_3"
},
{
"operation": "Highlighting",
"time": "60.356",
"value": "Gabriel René Paul_sentence_4"
},
{
"operation": "hyperlink",
"time": "68.775",
"value": "United States Military Academy"
},
{
"operation": "Now on",
"time": "70.916",
"value": "United States Military Academy"
},
{
"operation": "Highlighting",
"time": "75.614",
"value": "United States Military Academy_sentence_2"
},
{
"operation": "finish",
"time": "84.516",
"value": "finish"
}
]
}
|
26888
|
A friendship between three drummers led to the making of the album "Drumgasm".
|
{
"evidence": [
"Drumgasm_sentence_0",
"Drumgasm_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nA friendship between three drummers led to the making of the album \"Drumgasm\".\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Drumgasm_sentence_0",
"Drumgasm_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5735,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.516",
"value": "?search="
},
{
"operation": "search",
"time": "10.026",
"value": "Drumgasm"
},
{
"operation": "Now on",
"time": "11.15",
"value": "Drumgasm"
},
{
"operation": "Highlighting",
"time": "25.991",
"value": "Drumgasm_sentence_0"
},
{
"operation": "Highlighting",
"time": "28.725",
"value": "Drumgasm_sentence_2"
},
{
"operation": "finish",
"time": "45.455",
"value": "finish"
}
]
}
|
exfever_2094
|
Wayne Mark Rooney scored his first goal against Charlton Athletic in August 2003.
|
{
"evidence": [
"Wayne Rooney",
"EFL Championship"
],
"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:\nWayne Mark Rooney scored his first goal against Charlton Athletic in August 2003.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Wayne Rooney",
"EFL Championship"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5736,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7719
|
Lindsay Lohan is an American actress who appeared in the television soap opera that aired on FOX on May 4, 1964, which was created by Irna Phillips.
|
{
"evidence": [
"Lindsay Lohan",
"Another World (TV series)"
],
"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:\nLindsay Lohan is an American actress who appeared in the television soap opera that aired on FOX on May 4, 1964, which was created by Irna Phillips.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Lindsay Lohan",
"Another World (TV series)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5737,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
68746
|
The highest MAE % in the Gampaha Electoral District and Sri Lanka election was in 2005, while the lowest MAE % recorded was in the 1989 Gampaha Electoral District and Sri Lanka election.
|
{
"evidence": [
"Gampaha Electoral District_sentence_9",
"2005 Sri Lankan presidential election_sentence_0",
"Gampaha Electoral District_sentence_38"
],
"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 highest MAE % in the Gampaha Electoral District and Sri Lanka election was in 2005, while the lowest MAE % recorded was in the 1989 Gampaha Electoral District and Sri Lanka election.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Gampaha Electoral District_sentence_9",
"2005 Sri Lankan presidential election_sentence_0",
"Gampaha Electoral District_sentence_38"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5738,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.018",
"value": "?search="
},
{
"operation": "search",
"time": "18.223",
"value": "2005 Sri Lankan presidential election"
},
{
"operation": "Now on",
"time": "19.128",
"value": "2005 Sri Lankan presidential election"
},
{
"operation": "search",
"time": "41.667",
"value": "Gampaha Electoral District (1947–1989)"
},
{
"operation": "Now on",
"time": "43.591",
"value": "Gampaha Electoral District (1947–1989)"
},
{
"operation": "search",
"time": "86.642",
"value": "Gampaha Electoral District"
},
{
"operation": "Now on",
"time": "88.428",
"value": "Gampaha Electoral District"
},
{
"operation": "search",
"time": "117.586",
"value": "Contains...1989 Gampaha Electoral District and Sri Lanka election"
},
{
"operation": "Now on",
"time": "119.906",
"value": "undefined"
},
{
"operation": "search",
"time": "143.961",
"value": "Contains...1989 Gampaha Electoral District"
},
{
"operation": "Now on",
"time": "144.736",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "150.009",
"value": "Gampaha Electoral District (1947–1989)"
},
{
"operation": "Now on",
"time": "150.657",
"value": "Gampaha Electoral District (1947–1989)"
},
{
"operation": "search",
"time": "183.4",
"value": "1989 Sri Lankan parliamentary election"
},
{
"operation": "Now on",
"time": "184.562",
"value": "1989 Sri Lankan parliamentary election"
},
{
"operation": "hyperlink",
"time": "191.188",
"value": "Results of the 1989 Sri Lankan general election by electoral district"
},
{
"operation": "Now on",
"time": "194.643",
"value": "Results of the 1989 Sri Lankan general election by electoral district"
},
{
"operation": "back-button-clicked",
"time": "198.841",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "198.937",
"value": "1989 Sri Lankan parliamentary election"
},
{
"operation": "hyperlink",
"time": "200.253",
"value": "Results of the 1989 Sri Lankan general election by province"
},
{
"operation": "Now on",
"time": "201.845",
"value": "Results of the 1989 Sri Lankan general election by province"
},
{
"operation": "back-button-clicked",
"time": "230.055",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "230.148",
"value": "1989 Sri Lankan parliamentary election"
},
{
"operation": "hyperlink",
"time": "232.14",
"value": "Results of the 1989 Sri Lankan general election by electoral district"
},
{
"operation": "Now on",
"time": "232.75",
"value": "Results of the 1989 Sri Lankan general election by electoral district"
},
{
"operation": "hyperlink",
"time": "237.735",
"value": "Gampaha"
},
{
"operation": "Now on",
"time": "238.266",
"value": "Gampaha Electoral District"
},
{
"operation": "hyperlink",
"time": "243.411",
"value": "4 1989 parliamentary general election Gampaha Electoral District_section_3"
},
{
"operation": "hyperlink",
"time": "246.791",
"value": "9th parliamentary election"
},
{
"operation": "Now on",
"time": "247.264",
"value": "1989 Sri Lankan parliamentary election"
},
{
"operation": "back-button-clicked",
"time": "251.384",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "251.539",
"value": "Gampaha Electoral District#1989 parliamentary general election Gampaha Electoral District section 3"
},
{
"operation": "Highlighting",
"time": "255.657",
"value": "Gampaha Electoral District_sentence_9"
},
{
"operation": "hyperlink",
"time": "261.353",
"value": "9th parliamentary election"
},
{
"operation": "Now on",
"time": "261.87",
"value": "1989 Sri Lankan parliamentary election"
},
{
"operation": "hyperlink",
"time": "276.036",
"value": "Results of the 1989 Sri Lankan general election by electoral district"
},
{
"operation": "Now on",
"time": "276.797",
"value": "Results of the 1989 Sri Lankan general election by electoral district"
},
{
"operation": "hyperlink",
"time": "285.124",
"value": "Gampaha"
},
{
"operation": "Now on",
"time": "285.809",
"value": "Gampaha Electoral District"
},
{
"operation": "hyperlink",
"time": "296.21",
"value": "14 2005 presidential election Gampaha Electoral District_section_13"
},
{
"operation": "hyperlink",
"time": "299.276",
"value": "5th presidential election"
},
{
"operation": "Now on",
"time": "299.843",
"value": "2005 Sri Lankan presidential election"
},
{
"operation": "Highlighting",
"time": "315.581",
"value": "2005 Sri Lankan presidential election_sentence_0"
},
{
"operation": "back-button-clicked",
"time": "318.932",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "319.095",
"value": "Gampaha Electoral District#2005 presidential election Gampaha Electoral District section 13"
},
{
"operation": "Highlighting",
"time": "322.63",
"value": "Gampaha Electoral District_sentence_38"
},
{
"operation": "finish",
"time": "337.714",
"value": "finish"
}
]
}
|
28617
|
David Kranich is an American real estate agent and businessman. He ran for mayor of the District of Columbia in 2006, He ran for mayor of the District of Columbia in 2006.
|
{
"evidence": [
"David Kranich_sentence_0",
"David Kranich_sentence_1"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nDavid Kranich is an American real estate agent and businessman. He ran for mayor of the District of Columbia in 2006, He ran for mayor of the District of Columbia in 2006.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"David Kranich_sentence_0",
"David Kranich_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5739,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.196",
"value": "?search="
},
{
"operation": "search",
"time": "4.479",
"value": "David Kranich"
},
{
"operation": "Now on",
"time": "5.215",
"value": "David Kranich"
},
{
"operation": "Highlighting",
"time": "6.793",
"value": "David Kranich_sentence_0"
},
{
"operation": "Highlighting",
"time": "12.033",
"value": "David Kranich_sentence_1"
},
{
"operation": "finish",
"time": "23.557",
"value": "finish"
}
]
}
|
exfever_1591
|
The Punisher is a fictional antihero appearing in a thin book that averages 32 pages and contains comics that first gained popularity after the 1935 publication of Action Comics.
|
{
"evidence": [
"Punisher",
"American comic book"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Punisher is a fictional antihero appearing in a thin book that averages 32 pages and contains comics that first gained popularity after the 1935 publication of Action Comics.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Punisher",
"American comic book"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5740,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8336
|
Charlize Theron is a black American actress who rose to international prominence in the 1970s as the leading lady in Taylor Hackford's 1997 American supernatural horror film.
|
{
"evidence": [
"Charlize Theron",
"The Devil's Advocate (1997 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:\nCharlize Theron is a black American actress who rose to international prominence in the 1970s as the leading lady in Taylor Hackford's 1997 American supernatural horror film.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Charlize Theron",
"The Devil's Advocate (1997 film)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5741,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_254
|
My Summer of Love is the third film directed by Paweł Pawlikowski.
|
{
"evidence": [
"My Summer of Love",
"West Yorkshire"
],
"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:\nMy Summer of Love is the third film directed by Paweł Pawlikowski.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"My Summer of Love",
"West Yorkshire"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5742,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2204
|
Richard Nixon served five years in the White House, which witnessed the first arrival of a spacecraft on the surface of the Moon.
|
{
"evidence": [
"Richard Nixon",
"Moon landing"
],
"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:\nRichard Nixon served five years in the White House, which witnessed the first arrival of a spacecraft on the surface of the Moon.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Richard Nixon",
"Moon landing"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5743,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
15108
|
Born about 935, Elvira Ramírez was the daughter of the King Ramiro II of León (son of Ordoño II) and was made a nun by her father.
|
{
"evidence": [
"Elvira Ramírez_sentence_3",
"Elvira Ramírez_sentence_4",
"Ramiro II of León_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:\nBorn about 935, Elvira Ramírez was the daughter of the King Ramiro II of León (son of Ordoño II) and was made a nun by her father.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Elvira Ramírez_sentence_3",
"Elvira Ramírez_sentence_4",
"Ramiro II of León_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5744,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.584",
"value": "?search="
},
{
"operation": "search",
"time": "10.597",
"value": "Elvira Ramírez"
},
{
"operation": "Now on",
"time": "11.301",
"value": "Elvira Ramírez"
},
{
"operation": "Highlighting",
"time": "22.346",
"value": "Elvira Ramírez_sentence_3"
},
{
"operation": "Highlighting",
"time": "28.736",
"value": "Elvira Ramírez_sentence_4"
},
{
"operation": "hyperlink",
"time": "33.049",
"value": "Ramiro II of León"
},
{
"operation": "Now on",
"time": "33.795",
"value": "Ramiro II of León"
},
{
"operation": "Highlighting",
"time": "35.16",
"value": "Ramiro II of León_sentence_0"
},
{
"operation": "back-button-clicked",
"time": "36.408",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "36.898",
"value": "Elvira Ramírez"
},
{
"operation": "back-button-clicked",
"time": "37.031",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "37.048",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "37.207",
"value": "?search="
},
{
"operation": "finish",
"time": "47.568",
"value": "finish"
}
]
}
|
exfever_6359
|
Snoop Dogg was featured on Dr. Dre's debut solo single, 187, which was the debut solo single by American rapper Dr. Dre and his first track released after the breakup of N.W.A.
|
{
"evidence": [
"Snoop Dogg",
"Deep Cover (song)"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nSnoop Dogg was featured on Dr. Dre's debut solo single, 187, which was the debut solo single by American rapper Dr. Dre and his first track released after the breakup of N.W.A.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Snoop Dogg",
"Deep Cover (song)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5745,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
49653
|
Ricardian equivalence requires assumptions that have been seriously challenged; however, even in a laboratory setting where all assumptions required are ensured to hold, behaviour of individuals is consistent with Ricardian equivalence.
|
{
"evidence": [
"Ricardian equivalence_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:\nRicardian equivalence requires assumptions that have been seriously challenged; however, even in a laboratory setting where all assumptions required are ensured to hold, behaviour of individuals is consistent with Ricardian equivalence.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Ricardian equivalence_sentence_25"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5746,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.772",
"value": "?search="
},
{
"operation": "search",
"time": "7.852",
"value": "Ricardian equivalence"
},
{
"operation": "Now on",
"time": "8.863",
"value": "Ricardian equivalence"
},
{
"operation": "Highlighting",
"time": "29.218",
"value": "Ricardian equivalence_sentence_22"
},
{
"operation": "Highlighting",
"time": "33.887",
"value": "Ricardian equivalence_sentence_25"
},
{
"operation": "Highlighting deleted",
"time": "43.056",
"value": "Ricardian equivalence_sentence_22"
},
{
"operation": "finish",
"time": "56.522",
"value": "finish"
}
]
}
|
54761
|
Szymon Sawala (born September 24, 1982 in Sieraków) is am American footballer who plays as a guard.
|
{
"evidence": [
"Szymon Sawala_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:\nSzymon Sawala (born September 24, 1982 in Sieraków) is am American footballer who plays as a guard.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Szymon Sawala_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5747,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.073",
"value": "?search="
},
{
"operation": "search",
"time": "10.307",
"value": "Szymon Sawala"
},
{
"operation": "Now on",
"time": "11.37",
"value": "Szymon Sawala"
},
{
"operation": "Highlighting",
"time": "14.819",
"value": "Szymon Sawala_sentence_0"
},
{
"operation": "finish",
"time": "25.999",
"value": "finish"
}
]
}
|
exfever_8098
|
Chicago has 1 million residents living on the shore.
|
{
"evidence": [
"Chicago",
"Illinois",
"U.S. state"
],
"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:\nChicago has 1 million residents living on the shore.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Chicago",
"Illinois",
"U.S. state"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5748,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7067
|
A Song of Ice and Fire is the first epic fantasy novel by George R. R. Martin.
|
{
"evidence": [
"Bran Stark",
"A Feast for Crows"
],
"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:\nA Song of Ice and Fire is the first epic fantasy novel by George R. R. Martin.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Bran Stark",
"A Feast for Crows"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5749,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
72775
|
Martell Webster played for 12 seasons in the NBA.
|
{
"evidence": [
"Martell Webster_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:\nMartell Webster played for 12 seasons in the NBA.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Martell Webster_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5750,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.158",
"value": "?search="
},
{
"operation": "search",
"time": "8.833",
"value": "Martell Webster"
},
{
"operation": "Now on",
"time": "10.834",
"value": "Martell Webster"
},
{
"operation": "Highlighting",
"time": "16.543",
"value": "Martell Webster_sentence_0"
},
{
"operation": "finish",
"time": "31.531",
"value": "finish"
}
]
}
|
exfever_4214
|
The Taj Mahal is a black marble mausoleum on the right bank of the largest tributary river in India.
|
{
"evidence": [
"Taj Mahal",
"Yamuna"
],
"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 Taj Mahal is a black marble mausoleum on the right bank of the largest tributary river in India.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Taj Mahal",
"Yamuna"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5751,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
12360
|
Black Cat John Brown is the second album by Alamo Race Track which received positive ratings from De Volkskrant, Het Parool and Allmusic.
|
{
"evidence": [
"Black Cat John Brown_sentence_0",
"Black Cat John Brown_sentence_4"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBlack Cat John Brown is the second album by Alamo Race Track which received positive ratings from De Volkskrant, Het Parool and Allmusic.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Black Cat John Brown_sentence_0",
"Black Cat John Brown_sentence_4"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5752,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.639",
"value": "?search="
},
{
"operation": "search",
"time": "8.098",
"value": "Black Cat John Brown"
},
{
"operation": "Now on",
"time": "8.812",
"value": "Black Cat John Brown"
},
{
"operation": "Highlighting",
"time": "19.605",
"value": "Black Cat John Brown_sentence_0"
},
{
"operation": "Highlighting",
"time": "47.814",
"value": "Black Cat John Brown_sentence_4"
},
{
"operation": "hyperlink",
"time": "58.507",
"value": "De Volkskrant"
},
{
"operation": "Now on",
"time": "59.775",
"value": "De Volkskrant"
},
{
"operation": "back-button-clicked",
"time": "63.29",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "63.398",
"value": "Black Cat John Brown"
},
{
"operation": "hyperlink",
"time": "64.626",
"value": "Het Parool"
},
{
"operation": "Now on",
"time": "65.219",
"value": "Het Parool"
},
{
"operation": "back-button-clicked",
"time": "68.856",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "68.976",
"value": "Black Cat John Brown"
},
{
"operation": "hyperlink",
"time": "70.422",
"value": "Allmusic"
},
{
"operation": "Now on",
"time": "71.672",
"value": "Allmusic"
},
{
"operation": "back-button-clicked",
"time": "74.362",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "74.466",
"value": "Black Cat John Brown"
},
{
"operation": "finish",
"time": "93.47",
"value": "finish"
}
]
}
|
exfever_2938
|
The Assyrian people are a group of people who identify with each other on the basis of shared attributes that distinguish them from other groups located in a region in the Middle East.
|
{
"evidence": [
"Assyrian people",
"Ethnic group"
],
"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 Assyrian people are a group of people who identify with each other on the basis of shared attributes that distinguish them from other groups located in a region in the Middle East.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Assyrian people",
"Ethnic group"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5753,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
65387
|
John Lindsay set the world record for the Men's 200m TW3 at the 1992 Summer Paralympics (the ninth iteration of the event).
|
{
"evidence": [
"1992 Summer Paralympics_sentence_0",
"John Lindsay (Paralympian)_sentence_4"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJohn Lindsay set the world record for the Men's 200m TW3 at the 1992 Summer Paralympics (the ninth iteration of the event).\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"1992 Summer Paralympics_sentence_0",
"John Lindsay (Paralympian)_sentence_4"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5754,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.869",
"value": "?search="
},
{
"operation": "search",
"time": "12.499",
"value": "John Lindsay"
},
{
"operation": "Now on",
"time": "22.029",
"value": "John Lindsay"
},
{
"operation": "back-button-clicked",
"time": "42.275",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "42.377",
"value": "?search="
},
{
"operation": "search",
"time": "52.517",
"value": "1992 Summer Paralympics"
},
{
"operation": "Now on",
"time": "53.65",
"value": "1992 Summer Paralympics"
},
{
"operation": "Highlighting",
"time": "87.113",
"value": "1992 Summer Paralympics_sentence_0"
},
{
"operation": "search",
"time": "98.031",
"value": "Contains...Men's 200m TW3 at the 1992 Summer Paralympics"
},
{
"operation": "Now on",
"time": "100.414",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "127.174",
"value": "Canada at the 1992 Summer Paralympics"
},
{
"operation": "Now on",
"time": "128.554",
"value": "Canada at the 1992 Summer Paralympics"
},
{
"operation": "search",
"time": "169.902",
"value": "Contains...Men's 200m TW3"
},
{
"operation": "Now on",
"time": "200.137",
"value": "Canada at the 1992 Summer Paralympics"
},
{
"operation": "search",
"time": "215.105",
"value": "Men's 200m TW3"
},
{
"operation": "Now on",
"time": "216.106",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "220.243",
"value": "John Lindsay (Paralympian)"
},
{
"operation": "Now on",
"time": "220.692",
"value": "John Lindsay (Paralympian)"
},
{
"operation": "Highlighting",
"time": "230.647",
"value": "John Lindsay (Paralympian)_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "232.298",
"value": "John Lindsay (Paralympian)_sentence_2"
},
{
"operation": "Highlighting",
"time": "250.25",
"value": "John Lindsay (Paralympian)_sentence_4"
},
{
"operation": "hyperlink",
"time": "322.687",
"value": "1996 Summer Paralympics"
},
{
"operation": "Now on",
"time": "323.639",
"value": "1996 Summer Paralympics"
},
{
"operation": "back-button-clicked",
"time": "335.442",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "335.554",
"value": "John Lindsay (Paralympian)"
},
{
"operation": "finish",
"time": "371.332",
"value": "finish"
}
]
}
|
29349
|
The A647 road has nine major junctions in the city of Leeds, the cultural, financial and commercial heart of the West Yorkshire Built-up Area (2011 census classification).
|
{
"evidence": [
"A647 road_sentence_0",
"A647 road_sentence_2",
"A647 road_sentence_3",
"Leeds_sentence_4"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe A647 road has nine major junctions in the city of Leeds, the cultural, financial and commercial heart of the West Yorkshire Built-up Area (2011 census classification).\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"A647 road_sentence_0",
"A647 road_sentence_2",
"A647 road_sentence_3",
"Leeds_sentence_4"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5755,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.611",
"value": "?search="
},
{
"operation": "search",
"time": "125.415",
"value": "A647 road"
},
{
"operation": "Now on",
"time": "126.509",
"value": "A647 road"
},
{
"operation": "Highlighting",
"time": "149.882",
"value": "A647 road_sentence_0"
},
{
"operation": "Highlighting",
"time": "157.784",
"value": "A647 road_sentence_2"
},
{
"operation": "Highlighting",
"time": "159.321",
"value": "A647 road_sentence_3"
},
{
"operation": "hyperlink",
"time": "188.903",
"value": "Leeds"
},
{
"operation": "Now on",
"time": "192.072",
"value": "Leeds"
},
{
"operation": "Highlighting",
"time": "193.704",
"value": "Leeds_sentence_4"
},
{
"operation": "finish",
"time": "198.658",
"value": "finish"
}
]
}
|
exfever_6519
|
Minions is a 2015 American computer-animated drama film distributed by an American film production and distribution company owned by Comcast through the NBCUniversal Film and Entertainment division of NBCUniversal.
|
{
"evidence": [
"Minions (film)",
"Universal Pictures"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nMinions is a 2015 American computer-animated drama film distributed by an American film production and distribution company owned by Comcast through the NBCUniversal Film and Entertainment division of NBCUniversal.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Minions (film)",
"Universal Pictures"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5756,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3686
|
Holly Patricia Hunter won the award for best actress from AMPAS, that has been presenting the award since 1929, when they held the first ceremony on May 16 of the year at a private dinner held at the Hollywood Roosevelt Hotel in Los Angeles, California.
|
{
"evidence": [
"Holly Hunter",
"Academy Award for Best Actress",
"1st Academy Awards"
],
"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:\nHolly Patricia Hunter won the award for best actress from AMPAS, that has been presenting the award since 1929, when they held the first ceremony on May 16 of the year at a private dinner held at the Hollywood Roosevelt Hotel in Los Angeles, California.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Holly Hunter",
"Academy Award for Best Actress",
"1st Academy Awards"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5757,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
83303
|
EGR2 is an early growth response protein which is a group of amino acids, it is a transcription regulatory factor and is expressed in the neural crest derived cells of the cranial ganglion.
|
{
"evidence": [
"EGR2_sentence_0",
"Protein_sentence_3",
"EGR2_sentence_1",
"EGR2_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nEGR2 is an early growth response protein which is a group of amino acids, it is a transcription regulatory factor and is expressed in the neural crest derived cells of the cranial ganglion.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"EGR2_sentence_0",
"Protein_sentence_3",
"EGR2_sentence_1",
"EGR2_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5758,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.282",
"value": "?search="
},
{
"operation": "search",
"time": "13.889",
"value": "EGR2"
},
{
"operation": "Now on",
"time": "14.432",
"value": "EGR2"
},
{
"operation": "Highlighting",
"time": "36.123",
"value": "EGR2_sentence_0"
},
{
"operation": "hyperlink",
"time": "39.477",
"value": "protein"
},
{
"operation": "Now on",
"time": "41.213",
"value": "Protein"
},
{
"operation": "back-button-clicked",
"time": "47.369",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "47.432",
"value": "EGR2"
},
{
"operation": "hyperlink",
"time": "59.595",
"value": "protein"
},
{
"operation": "Now on",
"time": "60.08",
"value": "Protein"
},
{
"operation": "Highlighting",
"time": "64.452",
"value": "Protein_sentence_3"
},
{
"operation": "back-button-clicked",
"time": "188.81",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "188.876",
"value": "EGR2"
},
{
"operation": "Highlighting",
"time": "199.521",
"value": "EGR2_sentence_1"
},
{
"operation": "Highlighting",
"time": "211.163",
"value": "EGR2_sentence_2"
},
{
"operation": "hyperlink",
"time": "263.776",
"value": "protein"
},
{
"operation": "Now on",
"time": "264.393",
"value": "Protein"
},
{
"operation": "back-button-clicked",
"time": "269.457",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "269.517",
"value": "EGR2"
},
{
"operation": "finish",
"time": "361.779",
"value": "finish"
}
]
}
|
65512
|
Profilin 1 encoded by this gene is a ubiquitous actin and is thought to deregulate actin polymerization and it is associated with Miller-Dieker syndrome.
|
{
"evidence": [
"Profilin 1_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:\nProfilin 1 encoded by this gene is a ubiquitous actin and is thought to deregulate actin polymerization and it is associated with Miller-Dieker syndrome.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Profilin 1_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5759,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.501",
"value": "?search="
},
{
"operation": "search",
"time": "38.71",
"value": "Profilin 1 encoded b"
},
{
"operation": "Now on",
"time": "40.09",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "63.142",
"value": "PCLO"
},
{
"operation": "Now on",
"time": "64.068",
"value": "PCLO"
},
{
"operation": "search",
"time": "78.933",
"value": "Profilin 1"
},
{
"operation": "Now on",
"time": "79.966",
"value": "Profilin 1"
},
{
"operation": "Highlighting",
"time": "87.771",
"value": "Profilin 1_sentence_0"
},
{
"operation": "Highlighting",
"time": "111.665",
"value": "Profilin 1_sentence_1"
},
{
"operation": "Highlighting",
"time": "130.401",
"value": "Profilin 1_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "140.242",
"value": "Profilin 1_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "142.009",
"value": "Profilin 1_sentence_1"
},
{
"operation": "finish",
"time": "179.799",
"value": "finish"
}
]
}
|
91115
|
The albums "I Could Not Ask for More" and "Saints and Angels" were released in 2001.
|
{
"evidence": [
"I Could Not Ask for More_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 albums \"I Could Not Ask for More\" and \"Saints and Angels\" were released in 2001.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"I Could Not Ask for More_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5760,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.108",
"value": "?search="
},
{
"operation": "search",
"time": "8.903",
"value": "I Could Not Ask for More"
},
{
"operation": "Now on",
"time": "10.373",
"value": "I Could Not Ask for More"
},
{
"operation": "hyperlink",
"time": "12.093",
"value": "Diane Warren"
},
{
"operation": "Now on",
"time": "14.379",
"value": "Diane Warren"
},
{
"operation": "back-button-clicked",
"time": "25.332",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "25.488",
"value": "I Could Not Ask for More"
},
{
"operation": "Highlighting",
"time": "43.206",
"value": "I Could Not Ask for More_sentence_0"
},
{
"operation": "search",
"time": "49.984",
"value": "Saints and Angels"
},
{
"operation": "Now on",
"time": "53.744",
"value": "undefined"
},
{
"operation": "search",
"time": "60.077",
"value": "Saints and Angels album"
},
{
"operation": "Now on",
"time": "61.859",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "66.91",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "67.082",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "68.006",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "68.013",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "68.174",
"value": "I Could Not Ask for More"
},
{
"operation": "Page search",
"time": "71.295",
"value": "saints"
},
{
"operation": "search",
"time": "82.848",
"value": "\"Saints and Angels\""
},
{
"operation": "Now on",
"time": "83.659",
"value": "undefined"
},
{
"operation": "finish",
"time": "96.156",
"value": "finish"
}
]
}
|
exfever_4694
|
Meek Mill is an Indian rapper who signed with a record label imprint founded by Miami-based American rapper Rick Ross, which albums are distributed by an American record label founded in October 1947 by Ahmet Ertegun and Herb Abramson.
|
{
"evidence": [
"Meek Mill",
"Maybach Music Group",
"Atlantic Records"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nMeek Mill is an Indian rapper who signed with a record label imprint founded by Miami-based American rapper Rick Ross, which albums are distributed by an American record label founded in October 1947 by Ahmet Ertegun and Herb Abramson.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Meek Mill",
"Maybach Music Group",
"Atlantic Records"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5761,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_205
|
The number pi appears in many formulas across an area of knowledge that includes the study of such topics as numbers, quantity, and space, that may include a branch of mathematics dealing with limits and related theories.
|
{
"evidence": [
"Pi",
"Mathematics",
"Mathematical analysis"
],
"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 number pi appears in many formulas across an area of knowledge that includes the study of such topics as numbers, quantity, and space, that may include a branch of mathematics dealing with limits and related theories.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Pi",
"Mathematics",
"Mathematical analysis"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5762,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
88160
|
Richard Storrs Willis was impressed by Arthur Napoleão dos Santos, a Portuguese composer, pianist, instrument dealer and music publisher, when he performed in New York at age 15.
|
{
"evidence": [
"Arthur Napoleão dos Santos_sentence_1",
"Arthur Napoleão dos Santos_sentence_6",
"Arthur Napoleão dos Santos_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:\nRichard Storrs Willis was impressed by Arthur Napoleão dos Santos, a Portuguese composer, pianist, instrument dealer and music publisher, when he performed in New York at age 15.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Arthur Napoleão dos Santos_sentence_1",
"Arthur Napoleão dos Santos_sentence_6",
"Arthur Napoleão dos Santos_sentence_7"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5763,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.075",
"value": "?search="
},
{
"operation": "search",
"time": "13.411",
"value": "Arthur Napoleão dos Santos"
},
{
"operation": "Now on",
"time": "14.228",
"value": "Arthur Napoleão dos Santos"
},
{
"operation": "Highlighting",
"time": "25.06",
"value": "Arthur Napoleão dos Santos_sentence_1"
},
{
"operation": "Highlighting",
"time": "29.643",
"value": "Arthur Napoleão dos Santos_sentence_6"
},
{
"operation": "Highlighting",
"time": "35.852",
"value": "Arthur Napoleão dos Santos_sentence_7"
},
{
"operation": "finish",
"time": "44.522",
"value": "finish"
}
]
}
|
84116
|
John Puhl began his professional baseball career with the New London Whalers of the Connecticut State League in 1898 and played for the Whalers from May 4 until September 10.
|
{
"evidence": [
"John Puhl_sentence_1",
"John Puhl_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nJohn Puhl began his professional baseball career with the New London Whalers of the Connecticut State League in 1898 and played for the Whalers from May 4 until September 10.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"John Puhl_sentence_1",
"John Puhl_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5764,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.169",
"value": "?search="
},
{
"operation": "search",
"time": "12.877",
"value": "John Puhl"
},
{
"operation": "Now on",
"time": "13.775",
"value": "John Puhl"
},
{
"operation": "Highlighting",
"time": "23.296",
"value": "John Puhl_sentence_1"
},
{
"operation": "Highlighting",
"time": "28.479",
"value": "John Puhl_sentence_2"
},
{
"operation": "finish",
"time": "36.393",
"value": "finish"
}
]
}
|
59162
|
Lord Vishnu changed his appearance to be able to follow Aryamala.
|
{
"evidence": [
"Aryamala_sentence_19",
"Aryamala_sentence_23"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nLord Vishnu changed his appearance to be able to follow Aryamala.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Aryamala_sentence_19",
"Aryamala_sentence_23"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5765,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.959",
"value": "?search="
},
{
"operation": "search",
"time": "15.892",
"value": "Lord Vishnu"
},
{
"operation": "Now on",
"time": "19.046",
"value": "Vishnu"
},
{
"operation": "search",
"time": "83.607",
"value": "Lord Vishnu Temples in Kerala"
},
{
"operation": "Now on",
"time": "84.916",
"value": "List of Hindu temples in Kerala"
},
{
"operation": "search",
"time": "97.096",
"value": "Lord Vishnu aryamala"
},
{
"operation": "Now on",
"time": "97.736",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "99.99",
"value": "Aryamala"
},
{
"operation": "Now on",
"time": "100.558",
"value": "Aryamala"
},
{
"operation": "search",
"time": "162.237",
"value": "Lord Vishnu"
},
{
"operation": "Now on",
"time": "162.838",
"value": "Lord Vishnu"
},
{
"operation": "search",
"time": "230.6",
"value": "Aryamala"
},
{
"operation": "Now on",
"time": "231.065",
"value": "Aryamala"
},
{
"operation": "Highlighting",
"time": "253.126",
"value": "Aryamala_sentence_19"
},
{
"operation": "Highlighting",
"time": "258.277",
"value": "Aryamala_sentence_23"
},
{
"operation": "search",
"time": "355.191",
"value": "Prahalada (film)"
},
{
"operation": "Now on",
"time": "356.393",
"value": "Prahalada (film)"
},
{
"operation": "finish",
"time": "376.83",
"value": "finish"
}
]
}
|
exfever_377
|
Penny Dreadful's executive producers include a Barbadian stage director.
|
{
"evidence": [
"Penny Dreadful (TV series)",
"Sam Mendes"
],
"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:\nPenny Dreadful's executive producers include a Barbadian stage director.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Penny Dreadful (TV series)",
"Sam Mendes"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5766,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7753
|
Take Me High is a 1973 British feature film starring an American actor.
|
{
"evidence": [
"Take Me High",
"Anthony Andrews"
],
"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:\nTake Me High is a 1973 British feature film starring an American actor.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Take Me High",
"Anthony Andrews"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5767,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
52922
|
Mutations with the Ryanodine receptor 2 is linked to catecholaminergic polymorphic ventricular tachycardia which functions in the myocardium.
|
{
"evidence": [
"Ryanodine receptor 2_sentence_10",
"Ryanodine receptor 2_sentence_12"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nMutations with the Ryanodine receptor 2 is linked to catecholaminergic polymorphic ventricular tachycardia which functions in the myocardium.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Ryanodine receptor 2_sentence_10",
"Ryanodine receptor 2_sentence_12"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5768,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.121",
"value": "?search="
},
{
"operation": "search",
"time": "18.53",
"value": "Mutations with the Ryanodine receptor 2"
},
{
"operation": "Now on",
"time": "20.457",
"value": "undefined"
},
{
"operation": "search",
"time": "32.502",
"value": "Ryanodine receptor 2"
},
{
"operation": "Now on",
"time": "33.388",
"value": "Ryanodine receptor 2"
},
{
"operation": "Highlighting",
"time": "40.887",
"value": "Ryanodine receptor 2_sentence_10"
},
{
"operation": "hyperlink",
"time": "49.342",
"value": "catecholaminergic polymorphic ventricular tachycardia"
},
{
"operation": "Now on",
"time": "50.398",
"value": "Catecholaminergic polymorphic ventricular tachycardia"
},
{
"operation": "Page search",
"time": "71.225",
"value": "myocardium"
},
{
"operation": "back-button-clicked",
"time": "84.853",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "84.914",
"value": "Ryanodine receptor 2"
},
{
"operation": "Page search",
"time": "114.325",
"value": "myocardium"
},
{
"operation": "Highlighting",
"time": "119.036",
"value": "Ryanodine receptor 2_sentence_12"
},
{
"operation": "finish",
"time": "124.493",
"value": "finish"
}
]
}
|
74676
|
Before the release of the film Kannadasan, author N. R. Dasan accused Balachander of plagiarising Verum Mann which was eventually taken to the Madras High Court where Dasan won the case, where the judge ordered Balachander to pay him ₹1,000.
|
{
"evidence": [
"Apoorva Raagangal_sentence_55",
"Apoorva Raagangal_sentence_56",
"Apoorva Raagangal_sentence_57"
],
"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:\nBefore the release of the film Kannadasan, author N. R. Dasan accused Balachander of plagiarising Verum Mann which was eventually taken to the Madras High Court where Dasan won the case, where the judge ordered Balachander to pay him ₹1,000.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Apoorva Raagangal_sentence_55",
"Apoorva Raagangal_sentence_56",
"Apoorva Raagangal_sentence_57"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5769,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.533",
"value": "?search="
},
{
"operation": "search",
"time": "25.503",
"value": "Kannadasan"
},
{
"operation": "Now on",
"time": "26.508",
"value": "undefined"
},
{
"operation": "search",
"time": "70.65",
"value": "Before the release of the film Kannadasan, author N. R. Dasan accused Balachander of plagiarising Verum Mann"
},
{
"operation": "Now on",
"time": "73.527",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "76.061",
"value": "Apoorva Raagangal"
},
{
"operation": "Now on",
"time": "76.887",
"value": "Apoorva Raagangal"
},
{
"operation": "Page search",
"time": "86.098",
"value": "N. R. Dasan accused Balachander of plagiarising"
},
{
"operation": "Page search",
"time": "432.631",
"value": "N. R. Dasan accused Balachander of plagiarising"
},
{
"operation": "Highlighting",
"time": "437.751",
"value": "Apoorva Raagangal_sentence_55"
},
{
"operation": "Highlighting",
"time": "446.394",
"value": "Apoorva Raagangal_sentence_56"
},
{
"operation": "Highlighting",
"time": "448.616",
"value": "Apoorva Raagangal_sentence_57"
},
{
"operation": "finish",
"time": "466.808",
"value": "finish"
}
]
}
|
74949
|
Frances Farmer played as Melanie in Thunder Rock, a play that would later win the Peabody Award.
|
{
"evidence": [
"Frances Farmer_sentence_87"
],
"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:\nFrances Farmer played as Melanie in Thunder Rock, a play that would later win the Peabody Award.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Frances Farmer_sentence_87"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5770,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.84",
"value": "?search="
},
{
"operation": "search",
"time": "13.236",
"value": "Frances Farmer"
},
{
"operation": "Now on",
"time": "14.528",
"value": "Frances Farmer"
},
{
"operation": "Page search",
"time": "25.519",
"value": "thunder rock"
},
{
"operation": "Highlighting",
"time": "35.634",
"value": "Frances Farmer_sentence_87"
},
{
"operation": "Page search",
"time": "49.135",
"value": "peabody award"
},
{
"operation": "Page search",
"time": "86.464",
"value": "peabody"
},
{
"operation": "finish",
"time": "127.794",
"value": "finish"
}
]
}
|
exfever_723
|
Role Models is a 2008 American comedy film starring an American actor and comedian who appeared in a 2000 supernatural horror film directed by James Wong.
|
{
"evidence": [
"Role Models",
"Seann William Scott",
"Final Destination (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:\nRole Models is a 2008 American comedy film starring an American actor and comedian who appeared in a 2000 supernatural horror film directed by James Wong.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Role Models",
"Seann William Scott",
"Final Destination (film)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5771,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
88279
|
Australia exported gold solely to Europe in the past but the alluvial gold deposits were exhausted in the 1800s.
|
{
"evidence": [
"Australian gold rushes_sentence_264"
],
"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:\nAustralia exported gold solely to Europe in the past but the alluvial gold deposits were exhausted in the 1800s.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Australian gold rushes_sentence_264"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5772,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.06",
"value": "?search="
},
{
"operation": "search",
"time": "11.575",
"value": "Australia gold history"
},
{
"operation": "Now on",
"time": "14.45",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "20.788",
"value": "Gold Coast, Queensland"
},
{
"operation": "Now on",
"time": "22.782",
"value": "Gold Coast, Queensland"
},
{
"operation": "search",
"time": "49.824",
"value": "alluvial gold deposit Australia"
},
{
"operation": "Now on",
"time": "51.429",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "61.423",
"value": "Mining"
},
{
"operation": "Now on",
"time": "62.955",
"value": "Mining"
},
{
"operation": "back-button-clicked",
"time": "106.02",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "106.789",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "122.511",
"value": "Tamar Hematite Iron Company"
},
{
"operation": "Now on",
"time": "123.539",
"value": "Tamar Hematite Iron Company"
},
{
"operation": "search",
"time": "174.632",
"value": "gold import Europe"
},
{
"operation": "Now on",
"time": "176.266",
"value": "undefined"
},
{
"operation": "search",
"time": "219.207",
"value": "Australia Gold Industry"
},
{
"operation": "Now on",
"time": "220.679",
"value": "undefined"
},
{
"operation": "search",
"time": "271.169",
"value": "history of Australia"
},
{
"operation": "Now on",
"time": "276.434",
"value": "History of Australia"
},
{
"operation": "hyperlink",
"time": "313.786",
"value": "Australian gold rushes"
},
{
"operation": "Now on",
"time": "315.846",
"value": "Australian gold rushes"
},
{
"operation": "Highlighting",
"time": "342.509",
"value": "Australian gold rushes_sentence_264"
},
{
"operation": "search",
"time": "436.504",
"value": "list of Australian exports"
},
{
"operation": "Now on",
"time": "439.66",
"value": "undefined"
},
{
"operation": "search",
"time": "519.099",
"value": "economy of Australia"
},
{
"operation": "Now on",
"time": "521.355",
"value": "Economy of Australia"
},
{
"operation": "hyperlink",
"time": "575.823",
"value": "List of exports of Australia"
},
{
"operation": "Now on",
"time": "576.556",
"value": "List of exports of Australia"
},
{
"operation": "hyperlink",
"time": "584.418",
"value": "Gold"
},
{
"operation": "Now on",
"time": "586.827",
"value": "Gold"
},
{
"operation": "back-button-clicked",
"time": "615.706",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "615.807",
"value": "List of exports of Australia"
},
{
"operation": "hyperlink",
"time": "621.937",
"value": "Department of Foreign Affairs and Trade"
},
{
"operation": "Now on",
"time": "622.989",
"value": "Department of Foreign Affairs and Trade"
},
{
"operation": "search",
"time": "641.568",
"value": "exported gold solely to Europe"
},
{
"operation": "Now on",
"time": "643.158",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "692.724",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "693.686",
"value": "Department of Foreign Affairs and Trade"
},
{
"operation": "back-button-clicked",
"time": "696.582",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "696.586",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "696.634",
"value": "List of exports of Australia"
},
{
"operation": "hyperlink",
"time": "705.315",
"value": "Australia"
},
{
"operation": "Now on",
"time": "707.299",
"value": "Australia"
},
{
"operation": "back-button-clicked",
"time": "742.99",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "743.054",
"value": "List of exports of Australia"
},
{
"operation": "Highlighting",
"time": "746.218",
"value": "List of exports of Australia_cell_0_6_0"
},
{
"operation": "Highlighting",
"time": "746.22",
"value": "List of exports of Australia_title"
},
{
"operation": "Highlighting",
"time": "746.224",
"value": "List of exports of Australia_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "752.112",
"value": "List of exports of Australia_cell_0_6_0"
},
{
"operation": "Highlighting deleted",
"time": "752.114",
"value": "List of exports of Australia_title"
},
{
"operation": "Highlighting deleted",
"time": "752.117",
"value": "List of exports of Australia_header_cell_0_0_0"
},
{
"operation": "search",
"time": "758.922",
"value": "Austrailia"
},
{
"operation": "Now on",
"time": "761.217",
"value": "Australia"
},
{
"operation": "search",
"time": "769.55",
"value": "Australia Gold export"
},
{
"operation": "Now on",
"time": "770.566",
"value": "undefined"
},
{
"operation": "search",
"time": "827.455",
"value": "australian gold rush"
},
{
"operation": "Now on",
"time": "829.441",
"value": "Australian gold rushes"
},
{
"operation": "finish",
"time": "957.665",
"value": "finish"
}
]
}
|
83277
|
KCNR (1460 AM) is a television show broadcasting a Talk Radio format that emphasizes the importance of community focused radio.
|
{
"evidence": [
"KCNR_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:\nKCNR (1460 AM) is a television show broadcasting a Talk Radio format that emphasizes the importance of community focused radio.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"KCNR_sentence_3"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5773,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.833",
"value": "?search="
},
{
"operation": "search",
"time": "7.015",
"value": "KCNR"
},
{
"operation": "Now on",
"time": "8.013",
"value": "KCNR"
},
{
"operation": "Highlighting",
"time": "15.853",
"value": "KCNR_sentence_3"
},
{
"operation": "finish",
"time": "26.946",
"value": "finish"
}
]
}
|
71766
|
Rondeletiola minor, or the lentil bobtail, is a squid that grows to a mantle length of 23mm.
|
{
"evidence": [
"Rondeletiola minor_sentence_0",
"Rondeletiola minor_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nRondeletiola minor, or the lentil bobtail, is a squid that grows to a mantle length of 23mm.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Rondeletiola minor_sentence_0",
"Rondeletiola minor_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5774,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.693",
"value": "?search="
},
{
"operation": "search",
"time": "14.755",
"value": "Rondeletiola minor"
},
{
"operation": "Now on",
"time": "15.683",
"value": "Rondeletiola minor"
},
{
"operation": "Highlighting",
"time": "29.321",
"value": "Rondeletiola minor_sentence_0"
},
{
"operation": "Highlighting",
"time": "39.633",
"value": "Rondeletiola minor_sentence_2"
},
{
"operation": "finish",
"time": "60.343",
"value": "finish"
}
]
}
|
76859
|
The 1874 Baltimore Canaries season was a success as the team finished in first place winning 52 games.
|
{
"evidence": [
"1874 Baltimore Canaries season_sentence_1"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe 1874 Baltimore Canaries season was a success as the team finished in first place winning 52 games.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"1874 Baltimore Canaries season_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5775,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.394",
"value": "?search="
},
{
"operation": "search",
"time": "284.507",
"value": "1874 Baltimore Canaries season"
},
{
"operation": "Now on",
"time": "285.806",
"value": "1874 Baltimore Canaries season"
},
{
"operation": "Highlighting",
"time": "319.799",
"value": "1874 Baltimore Canaries season_sentence_1"
},
{
"operation": "Highlighting",
"time": "337.672",
"value": "1874 Baltimore Canaries season_cell_1_8_0"
},
{
"operation": "Highlighting",
"time": "337.674",
"value": "1874 Baltimore Canaries season_section_0"
},
{
"operation": "Highlighting",
"time": "337.674",
"value": "1874 Baltimore Canaries season_section_1"
},
{
"operation": "Highlighting",
"time": "337.677",
"value": "1874 Baltimore Canaries season_header_cell_1_0_0"
},
{
"operation": "Highlighting",
"time": "340.19",
"value": "1874 Baltimore Canaries season_cell_1_8_1"
},
{
"operation": "Highlighting",
"time": "340.193",
"value": "1874 Baltimore Canaries season_header_cell_1_0_1"
},
{
"operation": "Highlighting deleted",
"time": "355.651",
"value": "1874 Baltimore Canaries season_cell_1_8_0"
},
{
"operation": "Highlighting deleted",
"time": "355.657",
"value": "1874 Baltimore Canaries season_header_cell_1_0_0"
},
{
"operation": "Highlighting deleted",
"time": "356.17",
"value": "1874 Baltimore Canaries season_cell_1_8_1"
},
{
"operation": "Highlighting deleted",
"time": "356.173",
"value": "1874 Baltimore Canaries season_section_0"
},
{
"operation": "Highlighting deleted",
"time": "356.175",
"value": "1874 Baltimore Canaries season_section_1"
},
{
"operation": "Highlighting deleted",
"time": "356.178",
"value": "1874 Baltimore Canaries season_header_cell_1_0_1"
},
{
"operation": "finish",
"time": "370.849",
"value": "finish"
}
]
}
|
53514
|
George Lewis Scott, a mathematician and literary figure, began his literary career in London alongside James Thomson and Andrew Millar but opted to not participate in any educational initiatives.
|
{
"evidence": [
"George Lewis Scott_sentence_15"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nGeorge Lewis Scott, a mathematician and literary figure, began his literary career in London alongside James Thomson and Andrew Millar but opted to not participate in any educational initiatives.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"George Lewis Scott_sentence_15"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5776,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.513",
"value": "?search="
},
{
"operation": "search",
"time": "8.673",
"value": "George Lewis Scott"
},
{
"operation": "Now on",
"time": "9.997",
"value": "George Lewis Scott"
},
{
"operation": "Highlighting",
"time": "12.384",
"value": "George Lewis Scott_sentence_0"
},
{
"operation": "Highlighting",
"time": "29.449",
"value": "George Lewis Scott_sentence_14"
},
{
"operation": "Highlighting",
"time": "41.821",
"value": "George Lewis Scott_sentence_15"
},
{
"operation": "Highlighting deleted",
"time": "46.192",
"value": "George Lewis Scott_sentence_14"
},
{
"operation": "Highlighting deleted",
"time": "47.937",
"value": "George Lewis Scott_sentence_0"
},
{
"operation": "finish",
"time": "67.626",
"value": "finish"
}
]
}
|
exfever_2656
|
iO Theater was an improv theater that had notable alumni, including an American comedian who is also the co-founder of the improvisational-comedy troupe Upright Citizen Brigade.
|
{
"evidence": [
"IO Theater",
"Amy Poehler"
],
"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:\niO Theater was an improv theater that had notable alumni, including an American comedian who is also the co-founder of the improvisational-comedy troupe Upright Citizen Brigade.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"IO Theater",
"Amy Poehler"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 5777,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6914
|
The first female vice president of the United States is Kamala Harris.
|
{
"evidence": [
"Loretta Sanchez",
"Democratic Party (United States)"
],
"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 first female vice president of the United States is Kamala Harris.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Loretta Sanchez",
"Democratic Party (United States)"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5778,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4992
|
Liverpool is a professional soccer?club that plays with no one allowed to touch the ball except the major positions in association baseball.
|
{
"evidence": [
"Liverpool F.C.",
"Association football",
"Goalkeeper (association football)"
],
"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:\nLiverpool is a professional soccer?club that plays with no one allowed to touch the ball except the major positions in association baseball.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Liverpool F.C.",
"Association football",
"Goalkeeper (association football)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5779,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8495
|
The Last House on the Left is the first film by Dennis Iliadis.
|
{
"evidence": [
"The Last House on the Left (2009 film)",
"Horror 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:\nThe Last House on the Left is the first film by Dennis Iliadis.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Last House on the Left (2009 film)",
"Horror film"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5780,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
45314
|
Bedri Rahmi Eyüboğlu was born in Istanbul, Turkey and is the second child of a family with five.
|
{
"evidence": [
"Bedri Rahmi Eyüboğlu_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:\nBedri Rahmi Eyüboğlu was born in Istanbul, Turkey and is the second child of a family with five.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Bedri Rahmi Eyüboğlu_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5781,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.131",
"value": "?search="
},
{
"operation": "search",
"time": "10.796",
"value": "Bedri Rahmi Eyüboğlu"
},
{
"operation": "Now on",
"time": "11.644",
"value": "Bedri Rahmi Eyüboğlu"
},
{
"operation": "Highlighting",
"time": "243.584",
"value": "Bedri Rahmi Eyüboğlu_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "244.168",
"value": "Bedri Rahmi Eyüboğlu_sentence_0"
},
{
"operation": "Highlighting",
"time": "249.995",
"value": "Bedri Rahmi Eyüboğlu_sentence_1"
},
{
"operation": "finish",
"time": "263.788",
"value": "finish"
}
]
}
|
6437
|
Vratsa waterfalls(Skaklia, 141 metres (462 feet)and Borov Kamak, 63 metres) are waterfalls that overlook the town of Vratsa and the Vratsa lowlands and where remains from the ancient Bulgarian village Patleina were found
|
{
"evidence": [
"Vratsa waterfalls_sentence_0",
"Vratsa waterfalls_sentence_1",
"Vratsa waterfalls_sentence_3",
"Vratsa waterfalls_sentence_2",
"Vratsa waterfalls_sentence_8",
"Vratsa waterfalls_sentence_9",
"Vratsa waterfalls_sentence_4",
"Vratsa waterfalls_sentence_5"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nVratsa waterfalls(Skaklia, 141 metres (462 feet)and Borov Kamak, 63 metres) are waterfalls that overlook the town of Vratsa and the Vratsa lowlands and where remains from the ancient Bulgarian village Patleina were found\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Vratsa waterfalls_sentence_0",
"Vratsa waterfalls_sentence_1",
"Vratsa waterfalls_sentence_3",
"Vratsa waterfalls_sentence_2",
"Vratsa waterfalls_sentence_8",
"Vratsa waterfalls_sentence_9",
"Vratsa waterfalls_sentence_4",
"Vratsa waterfalls_sentence_5"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5782,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.891",
"value": "?search="
},
{
"operation": "search",
"time": "19.271",
"value": "Vratsa waterfalls"
},
{
"operation": "Now on",
"time": "20.742",
"value": "Vratsa waterfalls"
},
{
"operation": "Highlighting",
"time": "35.204",
"value": "Vratsa waterfalls_sentence_0"
},
{
"operation": "Highlighting",
"time": "38.211",
"value": "Vratsa waterfalls_sentence_1"
},
{
"operation": "Highlighting",
"time": "67.3",
"value": "Vratsa waterfalls_sentence_3"
},
{
"operation": "Highlighting",
"time": "73.715",
"value": "Vratsa waterfalls_sentence_2"
},
{
"operation": "Highlighting",
"time": "119.562",
"value": "Vratsa waterfalls_sentence_8"
},
{
"operation": "Page search",
"time": "178.607",
"value": "lowlands"
},
{
"operation": "Highlighting",
"time": "185.571",
"value": "Vratsa waterfalls_sentence_9"
},
{
"operation": "Highlighting",
"time": "216.561",
"value": "Vratsa waterfalls_sentence_4"
},
{
"operation": "Highlighting",
"time": "217.268",
"value": "Vratsa waterfalls_sentence_5"
},
{
"operation": "finish",
"time": "242.004",
"value": "finish"
}
]
}
|
exfever_7084
|
Adam Douglas Driver is the most famous American actor.
|
{
"evidence": [
"Adam Driver",
"Tony 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:\nAdam Douglas Driver is the most famous American actor.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Adam Driver",
"Tony Award"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 5783,
"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.