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 |
|---|---|---|---|---|---|---|---|---|
40806
|
Al Helfer and Rex Barney called the games for New York's WOR-TV; however, due to competition with games at Yankee Stadium, Phillies (finished the season in last place in the league) games were not able to match the audience for Yankee broadcasts.
|
{
"evidence": [
"Rex Barney_sentence_15",
"1958 Philadelphia Phillies season_sentence_7",
"1958 Philadelphia Phillies season_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:\nAl Helfer and Rex Barney called the games for New York's WOR-TV; however, due to competition with games at Yankee Stadium, Phillies (finished the season in last place in the league) games were not able to match the audience for Yankee broadcasts.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Rex Barney_sentence_15",
"1958 Philadelphia Phillies season_sentence_7",
"1958 Philadelphia Phillies season_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2484,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.302",
"value": "undefined"
},
{
"operation": "search",
"time": "22.863",
"value": "Al Helfer"
},
{
"operation": "Now on",
"time": "23.746",
"value": "Al Helfer"
},
{
"operation": "Page search",
"time": "36.847",
"value": "rex"
},
{
"operation": "search",
"time": "114.112",
"value": "wor-tv phillies"
},
{
"operation": "Now on",
"time": "115.184",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "134.116",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "134.348",
"value": "Al Helfer"
},
{
"operation": "Page search",
"time": "144.385",
"value": "wor"
},
{
"operation": "Page search",
"time": "162.597",
"value": "wor-tv"
},
{
"operation": "Highlighting",
"time": "178.093",
"value": "Al Helfer_sentence_14"
},
{
"operation": "Highlighting",
"time": "178.095",
"value": "Al Helfer_title"
},
{
"operation": "Highlighting",
"time": "178.096",
"value": "Al Helfer_section_0"
},
{
"operation": "search",
"time": "198.276",
"value": "Rex Barney"
},
{
"operation": "Now on",
"time": "199.122",
"value": "Rex Barney"
},
{
"operation": "Page search",
"time": "215.667",
"value": "wor-tv"
},
{
"operation": "Highlighting",
"time": "230.36",
"value": "Rex Barney_sentence_15"
},
{
"operation": "search",
"time": "296.939",
"value": "philadelphia phillies games broadcaast in new york"
},
{
"operation": "Now on",
"time": "298.808",
"value": "undefined"
},
{
"operation": "search",
"time": "315.976",
"value": "wor-tv phillies games"
},
{
"operation": "Now on",
"time": "316.63",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "336.445",
"value": "Mutual Broadcasting System"
},
{
"operation": "Now on",
"time": "337.318",
"value": "Mutual Broadcasting System"
},
{
"operation": "Page search",
"time": "347.212",
"value": "phillies"
},
{
"operation": "Page search",
"time": "376.107",
"value": "yankee"
},
{
"operation": "Page search",
"time": "422.851",
"value": "phillies"
},
{
"operation": "search",
"time": "448.83",
"value": "due to competition with games at Yankee Stadium, Phillies (finished the season in last place in the league) games were not able to match the audience for Yankee broadcasts."
},
{
"operation": "Now on",
"time": "451.677",
"value": "undefined"
},
{
"operation": "search",
"time": "493.035",
"value": "hillies games were not able to match the audience for Yankee broadcasts."
},
{
"operation": "Now on",
"time": "493.591",
"value": "undefined"
},
{
"operation": "search",
"time": "534.918",
"value": "1958 Philadelphia Phillies season"
},
{
"operation": "Now on",
"time": "536.303",
"value": "1958 Philadelphia Phillies season"
},
{
"operation": "Page search",
"time": "541.474",
"value": "helfer"
},
{
"operation": "Highlighting",
"time": "553.632",
"value": "1958 Philadelphia Phillies season_sentence_7"
},
{
"operation": "Page search",
"time": "579.422",
"value": "last"
},
{
"operation": "Highlighting",
"time": "584.054",
"value": "1958 Philadelphia Phillies season_sentence_1"
},
{
"operation": "finish",
"time": "600.032",
"value": "finish"
}
]
}
|
93366
|
The TPS is a synchrotron light source project launched in 2007.
|
{
"evidence": [
"National Synchrotron Radiation Research Center_sentence_6",
"National Synchrotron Radiation Research Center_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:\nThe TPS is a synchrotron light source project launched in 2007.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"National Synchrotron Radiation Research Center_sentence_6",
"National Synchrotron Radiation Research Center_sentence_7"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2485,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.74",
"value": "?search="
},
{
"operation": "search",
"time": "6.023",
"value": "TPS"
},
{
"operation": "Now on",
"time": "6.543",
"value": "TPS"
},
{
"operation": "search",
"time": "78.025",
"value": "The TPS is a synchrotron light source project"
},
{
"operation": "Now on",
"time": "79.45",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "81",
"value": "National Synchrotron Radiation Research Center"
},
{
"operation": "Now on",
"time": "81.796",
"value": "National Synchrotron Radiation Research Center"
},
{
"operation": "Highlighting",
"time": "101.232",
"value": "National Synchrotron Radiation Research Center_sentence_7"
},
{
"operation": "Highlighting deleted",
"time": "102.024",
"value": "National Synchrotron Radiation Research Center_sentence_7"
},
{
"operation": "Highlighting",
"time": "103.362",
"value": "National Synchrotron Radiation Research Center_sentence_6"
},
{
"operation": "Highlighting",
"time": "106.436",
"value": "National Synchrotron Radiation Research Center_sentence_7"
},
{
"operation": "finish",
"time": "141.838",
"value": "finish"
}
]
}
|
exfever_7224
|
Jennifer Jason Leigh was nominated for Best Actress in a Motion Picture – Drama at the Golden Globe for her portrayal of Dorothy Parker.
|
{
"evidence": [
"Jennifer Jason Leigh",
"Miami Blues"
],
"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:\nJennifer Jason Leigh was nominated for Best Actress in a Motion Picture – Drama at the Golden Globe for her portrayal of Dorothy Parker.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Jennifer Jason Leigh",
"Miami Blues"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2486,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_89
|
The Great Buck Howard is an American comedy-drama film starring an American filmmaker known for his comedic and dramatic roles.
|
{
"evidence": [
"The Great Buck Howard",
"Tom Hanks"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Great Buck Howard is an American comedy-drama film starring an American filmmaker known for his comedic and dramatic roles.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Great Buck Howard",
"Tom Hanks"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2487,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
27021
|
Mikhail Ignatov made his debut in the Russian Football National League for FC Spartak-2 Moscow, he played in Russian Premier League and scored his first goal against FC Yenisey Krasnoyarsk which was founded in 1937.
|
{
"evidence": [
"Mikhail Ignatov_sentence_3",
"Mikhail Ignatov_sentence_5",
"Mikhail Ignatov_sentence_6",
"FC Yenisey Krasnoyarsk_sentence_3"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nMikhail Ignatov made his debut in the Russian Football National League for FC Spartak-2 Moscow, he played in Russian Premier League and scored his first goal against FC Yenisey Krasnoyarsk which was founded in 1937.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Mikhail Ignatov_sentence_3",
"Mikhail Ignatov_sentence_5",
"Mikhail Ignatov_sentence_6",
"FC Yenisey Krasnoyarsk_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2488,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "30.241",
"value": "undefined"
},
{
"operation": "search",
"time": "46.681",
"value": "Mikhail Ignatov"
},
{
"operation": "Now on",
"time": "47.951",
"value": "Mikhail Ignatov"
},
{
"operation": "Highlighting",
"time": "52.268",
"value": "Mikhail Ignatov_sentence_1"
},
{
"operation": "Highlighting",
"time": "57.213",
"value": "Mikhail Ignatov_sentence_3"
},
{
"operation": "Highlighting",
"time": "87.78",
"value": "Mikhail Ignatov_sentence_5"
},
{
"operation": "Highlighting",
"time": "100.828",
"value": "Mikhail Ignatov_sentence_6"
},
{
"operation": "Highlighting deleted",
"time": "106.456",
"value": "Mikhail Ignatov_sentence_1"
},
{
"operation": "hyperlink",
"time": "109.925",
"value": "FC Yenisey Krasnoyarsk"
},
{
"operation": "Now on",
"time": "111.034",
"value": "FC Yenisey Krasnoyarsk"
},
{
"operation": "Highlighting",
"time": "116.776",
"value": "FC Yenisey Krasnoyarsk_sentence_3"
},
{
"operation": "finish",
"time": "124.214",
"value": "finish"
}
]
}
|
77413
|
What Dreams May Come, a 2018 American drama film written and directed by Jennifer Fox, was nominated for three categories in the 2018 Independent Spirit Award.
|
{
"evidence": [
"What Dreams May Come (film)_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:\nWhat Dreams May Come, a 2018 American drama film written and directed by Jennifer Fox, was nominated for three categories in the 2018 Independent Spirit Award.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"What Dreams May Come (film)_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2489,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.378",
"value": "?search="
},
{
"operation": "search",
"time": "22.12",
"value": "What Dreams May Come (film)"
},
{
"operation": "Now on",
"time": "23.295",
"value": "What Dreams May Come (film)"
},
{
"operation": "search",
"time": "90.631",
"value": "What Dreams May Come"
},
{
"operation": "Now on",
"time": "91.936",
"value": "What Dreams May Come"
},
{
"operation": "search",
"time": "118.715",
"value": "Jennifer Fox"
},
{
"operation": "Now on",
"time": "119.808",
"value": "Jennifer Fox"
},
{
"operation": "hyperlink",
"time": "123.94",
"value": "Jennifer Fox (film producer)"
},
{
"operation": "Now on",
"time": "124.523",
"value": "Jennifer Fox (film producer)"
},
{
"operation": "search",
"time": "213.497",
"value": "What Dreams May Come"
},
{
"operation": "Now on",
"time": "214.454",
"value": "What Dreams May Come"
},
{
"operation": "Highlighting",
"time": "220.948",
"value": "What Dreams May Come_sentence_1"
},
{
"operation": "Highlighting",
"time": "220.95",
"value": "What Dreams May Come_title"
},
{
"operation": "Highlighting deleted",
"time": "223.401",
"value": "What Dreams May Come_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "223.405",
"value": "What Dreams May Come_title"
},
{
"operation": "hyperlink",
"time": "224.87",
"value": "What Dreams May Come (film)"
},
{
"operation": "Now on",
"time": "225.783",
"value": "What Dreams May Come (film)"
},
{
"operation": "Highlighting",
"time": "228.038",
"value": "What Dreams May Come (film)_sentence_0"
},
{
"operation": "Highlighting",
"time": "250.579",
"value": "What Dreams May Come (film)_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "251.775",
"value": "What Dreams May Come (film)_cell_0_1_1"
},
{
"operation": "Highlighting",
"time": "251.788",
"value": "What Dreams May Come (film)_header_cell_0_1_0"
},
{
"operation": "Highlighting",
"time": "251.788",
"value": "What Dreams May Come (film)_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "258.198",
"value": "What Dreams May Come (film)_header_cell_0_3_0"
},
{
"operation": "Highlighting",
"time": "259.356",
"value": "What Dreams May Come (film)_cell_0_3_1"
},
{
"operation": "Highlighting",
"time": "259.372",
"value": "What Dreams May Come (film)_header_cell_0_3_0"
},
{
"operation": "Highlighting",
"time": "259.373",
"value": "What Dreams May Come (film)_header_cell_0_0_0"
},
{
"operation": "finish",
"time": "337.701",
"value": "finish"
}
]
}
|
56483
|
In July 2015, the Council approved plans for a $500 million redevelopment of the centre; then on March 15, 2017, plans for the upgrade were unveiled.
|
{
"evidence": [
"The Glen Shopping Centre_sentence_11",
"The Glen Shopping Centre_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:\nIn July 2015, the Council approved plans for a $500 million redevelopment of the centre; then on March 15, 2017, plans for the upgrade were unveiled.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"The Glen Shopping Centre_sentence_11",
"The Glen Shopping Centre_sentence_12"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2490,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.935",
"value": "?search="
},
{
"operation": "search",
"time": "30.348",
"value": "Council approved plans for a $500 million redevelopment"
},
{
"operation": "Now on",
"time": "33.206",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "45.568",
"value": "The Glen Shopping Centre"
},
{
"operation": "Now on",
"time": "46.189",
"value": "The Glen Shopping Centre"
},
{
"operation": "Highlighting",
"time": "53.447",
"value": "The Glen Shopping Centre_sentence_11"
},
{
"operation": "Highlighting",
"time": "58.412",
"value": "The Glen Shopping Centre_sentence_12"
},
{
"operation": "finish",
"time": "89.819",
"value": "finish"
}
]
}
|
exfever_1912
|
Ghostbusters is a 2016 American supernatural comedy film starring an American actress, comedian, and writer who appeared in a 2014 American live-action/animated sports comedy film.
|
{
"evidence": [
"Ghostbusters (2016 film)",
"Kate McKinnon",
"Balls Out (2014 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:\nGhostbusters is a 2016 American supernatural comedy film starring an American actress, comedian, and writer who appeared in a 2014 American live-action/animated sports comedy film.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Ghostbusters (2016 film)",
"Kate McKinnon",
"Balls Out (2014 film)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2491,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8972
|
Logic is an American rapper who released his debut album on October 21, 2014 via Def Jam Recordings.
|
{
"evidence": [
"Logic (musician)",
"Under Pressure (album)"
],
"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:\nLogic is an American rapper who released his debut album on October 21, 2014 via Def Jam Recordings.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Logic (musician)",
"Under Pressure (album)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2492,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
83979
|
In the 1994 European Grand Prix, a Formula One motor race held on 16 October 1994 at the Circuito Permanente de Jerez, Jerez, Spain, Michael Schumacher placed second.
|
{
"evidence": [
"1994 European Grand Prix_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:\nIn the 1994 European Grand Prix, a Formula One motor race held on 16 October 1994 at the Circuito Permanente de Jerez, Jerez, Spain, Michael Schumacher placed second.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"1994 European Grand Prix_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2493,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.186",
"value": "?search="
},
{
"operation": "search",
"time": "12.883",
"value": "1994 European Grand Prix"
},
{
"operation": "Now on",
"time": "14.328",
"value": "1994 European Grand Prix"
},
{
"operation": "Highlighting",
"time": "23.122",
"value": "1994 European Grand Prix_sentence_0"
},
{
"operation": "Highlighting",
"time": "37.824",
"value": "1994 European Grand Prix_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "40.332",
"value": "1994 European Grand Prix_sentence_0"
},
{
"operation": "finish",
"time": "51.738",
"value": "finish"
}
]
}
|
20490
|
In 1982 United States presidential election in Tennessee, Richard Nixon (38th president of the United States) got the highest votes of 713,147, while John G. Schmitz got the lowest votes of 10,373.
|
{
"evidence": [
"Richard Nixon_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:\nIn 1982 United States presidential election in Tennessee, Richard Nixon (38th president of the United States) got the highest votes of 713,147, while John G. Schmitz got the lowest votes of 10,373.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Richard Nixon_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2494,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.153",
"value": "?search="
},
{
"operation": "search",
"time": "17.584",
"value": "1982 United States presidential election in Tennessee"
},
{
"operation": "Now on",
"time": "20.549",
"value": "undefined"
},
{
"operation": "search",
"time": "29.314",
"value": "1982 United States presidential election Tennessee"
},
{
"operation": "Now on",
"time": "30.081",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "34.309",
"value": "2016 United States presidential election"
},
{
"operation": "Now on",
"time": "36.552",
"value": "2016 United States presidential election"
},
{
"operation": "Page search",
"time": "43.059",
"value": "Tennessee"
},
{
"operation": "Highlighting",
"time": "71.295",
"value": "2016 United States presidential election_sentence_82"
},
{
"operation": "Highlighting",
"time": "71.297",
"value": "2016 United States presidential election_title"
},
{
"operation": "Highlighting",
"time": "71.299",
"value": "2016 United States presidential election_section_2"
},
{
"operation": "Highlighting",
"time": "71.299",
"value": "2016 United States presidential election_section_8"
},
{
"operation": "Highlighting",
"time": "71.3",
"value": "2016 United States presidential election_section_9"
},
{
"operation": "Highlighting deleted",
"time": "74.679",
"value": "2016 United States presidential election_sentence_82"
},
{
"operation": "Highlighting deleted",
"time": "74.683",
"value": "2016 United States presidential election_title"
},
{
"operation": "Highlighting deleted",
"time": "74.686",
"value": "2016 United States presidential election_section_2"
},
{
"operation": "Highlighting deleted",
"time": "74.688",
"value": "2016 United States presidential election_section_8"
},
{
"operation": "Highlighting deleted",
"time": "74.691",
"value": "2016 United States presidential election_section_9"
},
{
"operation": "search",
"time": "81.912",
"value": "Contains...1982 United States presidential election in Tennessee"
},
{
"operation": "Now on",
"time": "82.955",
"value": "undefined"
},
{
"operation": "search",
"time": "129.692",
"value": "1992 United States presidential election in Tennessee"
},
{
"operation": "Now on",
"time": "131.304",
"value": "1992 United States presidential election in Tennessee"
},
{
"operation": "search",
"time": "192.149",
"value": "1982 United States presidential election in Tennessee"
},
{
"operation": "Now on",
"time": "194.178",
"value": "undefined"
},
{
"operation": "search",
"time": "217.713",
"value": "RIchard Nixon presidential election"
},
{
"operation": "Now on",
"time": "219.454",
"value": "undefined"
},
{
"operation": "search",
"time": "236.748",
"value": "Richard Nixon"
},
{
"operation": "Now on",
"time": "238.746",
"value": "Richard Nixon"
},
{
"operation": "Highlighting",
"time": "244.422",
"value": "Richard Nixon_sentence_2"
},
{
"operation": "finish",
"time": "257.499",
"value": "finish"
}
]
}
|
exfever_6699
|
Tropic Thunder is a 2008 action-comedy film directed by Ben Stiller, who wrote the screenplay with an American actor and filmmaker born on August 10, 1971.
|
{
"evidence": [
"Tropic Thunder",
"Justin Theroux"
],
"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 directed by Ben Stiller, who wrote the screenplay with an American actor and filmmaker born on August 10, 1971.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tropic Thunder",
"Justin Theroux"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2495,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2889
|
Christianity is the first religion of the Assyrian people.
|
{
"evidence": [
"Assyrian people",
"Mesopotamia"
],
"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:\nChristianity is the first religion of the Assyrian people.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Assyrian people",
"Mesopotamia"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2496,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
17346
|
In 1963, Gam was a leading member of The Minnesota Theatre Company in the opening season of The Tyrone Guthrie Theater in Minneapolis; then in 2003, she appeared in the rotating cast of the Off-Broadway stage reading of Wit & Wisdom.
|
{
"evidence": [
"Rita Gam_sentence_12",
"Rita Gam_sentence_15"
],
"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 1963, Gam was a leading member of The Minnesota Theatre Company in the opening season of The Tyrone Guthrie Theater in Minneapolis; then in 2003, she appeared in the rotating cast of the Off-Broadway stage reading of Wit & Wisdom.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Rita Gam_sentence_12",
"Rita Gam_sentence_15"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2497,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.696",
"value": "?search="
},
{
"operation": "search",
"time": "22.154",
"value": "Gam"
},
{
"operation": "Now on",
"time": "24.467",
"value": "Gam"
},
{
"operation": "hyperlink",
"time": "42.306",
"value": "Rita Gam"
},
{
"operation": "Now on",
"time": "43.582",
"value": "Rita Gam"
},
{
"operation": "Page search",
"time": "55.329",
"value": "The Tyrone Guthrie Theater"
},
{
"operation": "Highlighting",
"time": "59.483",
"value": "Rita Gam_sentence_12"
},
{
"operation": "Highlighting",
"time": "70.502",
"value": "Rita Gam_sentence_15"
},
{
"operation": "finish",
"time": "87.558",
"value": "finish"
}
]
}
|
exfever_8559
|
Lindsey Buckingham is a music director.
|
{
"evidence": [
"Stevie Nicks",
"Fleetwood Mac"
],
"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:\nLindsey Buckingham is a music director.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Stevie Nicks",
"Fleetwood Mac"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2498,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
84314
|
Prince Semyon Prozorovsky led a detachment of tsarist troops to defend the Tikhvin Monastery against a 5000 man army of the Swedes.
|
{
"evidence": [
"Siege of Tikhvin_sentence_4",
"Siege of Tikhvin_sentence_5"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nPrince Semyon Prozorovsky led a detachment of tsarist troops to defend the Tikhvin Monastery against a 5000 man army of the Swedes.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Siege of Tikhvin_sentence_4",
"Siege of Tikhvin_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2499,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.419",
"value": "?search="
},
{
"operation": "search",
"time": "59.032",
"value": "semyon prozorovsky"
},
{
"operation": "Now on",
"time": "60.308",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "70.793",
"value": "George Learmonth"
},
{
"operation": "Now on",
"time": "71.642",
"value": "George Learmonth"
},
{
"operation": "hyperlink",
"time": "82.058",
"value": "Prozorovsky"
},
{
"operation": "Now on",
"time": "82.64",
"value": "Prozorovsky"
},
{
"operation": "search",
"time": "175.18",
"value": "prince semyonovich prozorovsky"
},
{
"operation": "Now on",
"time": "175.743",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "187.071",
"value": "List of Russian princely families"
},
{
"operation": "Now on",
"time": "189.159",
"value": "List of Russian princely families"
},
{
"operation": "search",
"time": "222.766",
"value": "detachment of tsarist troops to defend the Tikhvin Monastery"
},
{
"operation": "Now on",
"time": "224.104",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "228.761",
"value": "Siege of the Tikhvin Monastery Siege of Tikhvin_section_1"
},
{
"operation": "Now on",
"time": "229.198",
"value": "Siege of Tikhvin#Siege of the Tikhvin Monastery Siege of Tikhvin section 1"
},
{
"operation": "Highlighting",
"time": "245.35",
"value": "Siege of Tikhvin_sentence_4"
},
{
"operation": "Highlighting",
"time": "259.93",
"value": "Siege of Tikhvin_sentence_5"
},
{
"operation": "finish",
"time": "310.913",
"value": "finish"
}
]
}
|
75786
|
Kai Krause (named one of the top 15 innovators in 2005) failed to move to California in 1978 and worked with early synthesizers to make sound effects.
|
{
"evidence": [
"Kai Krause_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:\nKai Krause (named one of the top 15 innovators in 2005) failed to move to California in 1978 and worked with early synthesizers to make sound effects.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Kai Krause_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2500,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.272",
"value": "?search="
},
{
"operation": "search",
"time": "7.008",
"value": "Kai Krause"
},
{
"operation": "Now on",
"time": "7.8",
"value": "Kai Krause"
},
{
"operation": "Highlighting",
"time": "96.352",
"value": "Kai Krause_sentence_4"
},
{
"operation": "Highlighting",
"time": "145.231",
"value": "Kai Krause_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "148.494",
"value": "Kai Krause_sentence_4"
},
{
"operation": "finish",
"time": "161.264",
"value": "finish"
}
]
}
|
exfever_8258
|
The Marvel Cinematic Universe centered on a series of superhero films that are based on characters that appear in a thick book, averaging 32 pages and containing comics published by Marvel Comics.
|
{
"evidence": [
"Marvel Cinematic Universe",
"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 Marvel Cinematic Universe centered on a series of superhero films that are based on characters that appear in a thick book, averaging 32 pages and containing comics published by Marvel Comics.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Marvel Cinematic Universe",
"American comic book"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2501,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6807
|
The Fame is Lady Gaga's debut studio album, which was produced by different producers, including Cherry Cherry Boom Boom, a British music personality.
|
{
"evidence": [
"The Fame",
"Martin Kierszenbaum"
],
"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 Fame is Lady Gaga's debut studio album, which was produced by different producers, including Cherry Cherry Boom Boom, a British music personality.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Fame",
"Martin Kierszenbaum"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2502,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5023
|
Angelina Jolie is an American actress who starred in a 1997 biographical two-part mini-series produced and directed by John Frankenheimer.
|
{
"evidence": [
"Angelina Jolie",
"George Wallace (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:\nAngelina Jolie is an American actress who starred in a 1997 biographical two-part mini-series produced and directed by John Frankenheimer.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Angelina Jolie",
"George Wallace (film)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2503,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
24020
|
In the 1996 United States Senate election Texas Republican Phil Gramm won over Democrat Victor M. Morales.
|
{
"evidence": [
"1996 United States Senate elections_sentence_304",
"Victor Morales (politician)_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:\nIn the 1996 United States Senate election Texas Republican Phil Gramm won over Democrat Victor M. Morales.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"1996 United States Senate elections_sentence_304",
"Victor Morales (politician)_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2504,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.589",
"value": "?search="
},
{
"operation": "search",
"time": "18.415",
"value": "1996 United States Senate election"
},
{
"operation": "Now on",
"time": "23.206",
"value": "1996 United States Senate elections"
},
{
"operation": "hyperlink",
"time": "34.621",
"value": "35 Texas 1996 United States Senate elections_section_45"
},
{
"operation": "Highlighting",
"time": "47.071",
"value": "1996 United States Senate elections_sentence_304"
},
{
"operation": "hyperlink",
"time": "66.265",
"value": "Victor Morales"
},
{
"operation": "Now on",
"time": "66.804",
"value": "Victor Morales (politician)"
},
{
"operation": "Highlighting",
"time": "68.462",
"value": "Victor Morales (politician)_sentence_0"
},
{
"operation": "finish",
"time": "75.332",
"value": "finish"
}
]
}
|
8751
|
Konstantin Feoktistov joined the U.S Army at the age of 21, shot an Afghanistan officer and promoted by the U.S Army.
|
{
"evidence": [
"Konstantin Feoktistov_sentence_6"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nKonstantin Feoktistov joined the U.S Army at the age of 21, shot an Afghanistan officer and promoted by the U.S Army.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Konstantin Feoktistov_sentence_6"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2505,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.718",
"value": "?search="
},
{
"operation": "search",
"time": "15.374",
"value": "Konstantin Feoktistov"
},
{
"operation": "Now on",
"time": "17.046",
"value": "Konstantin Feoktistov"
},
{
"operation": "Highlighting",
"time": "48.2",
"value": "Konstantin Feoktistov_sentence_6"
},
{
"operation": "Highlighting",
"time": "84.441",
"value": "Konstantin Feoktistov_header_cell_0_4_0"
},
{
"operation": "Highlighting deleted",
"time": "91.181",
"value": "Konstantin Feoktistov_header_cell_0_4_0"
},
{
"operation": "hyperlink",
"time": "131.137",
"value": "Feoktistov (disambiguation)"
},
{
"operation": "Now on",
"time": "132.284",
"value": "Feoktistov (disambiguation)"
},
{
"operation": "finish",
"time": "146.067",
"value": "finish"
}
]
}
|
35170
|
Hosokawa Yoriyuki was the Shogun's Deputy in 1367, and he served as Yoshimitsu's historian for six years
|
{
"evidence": [
"Hosokawa Yoriyuki_sentence_1",
"Hosokawa Yoriyuki_sentence_9"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nHosokawa Yoriyuki was the Shogun's Deputy in 1367, and he served as Yoshimitsu's historian for six years\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Hosokawa Yoriyuki_sentence_1",
"Hosokawa Yoriyuki_sentence_9"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2506,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.779",
"value": "?search="
},
{
"operation": "search",
"time": "14.245",
"value": "Hosokawa Yoriyuki"
},
{
"operation": "Now on",
"time": "15.341",
"value": "Hosokawa Yoriyuki"
},
{
"operation": "Highlighting",
"time": "24.383",
"value": "Hosokawa Yoriyuki_sentence_1"
},
{
"operation": "Page search",
"time": "38.832",
"value": "historian"
},
{
"operation": "search",
"time": "130.127",
"value": "Yoshimitsu"
},
{
"operation": "Now on",
"time": "131.561",
"value": "Yoshimitsu"
},
{
"operation": "back-button-clicked",
"time": "163.806",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "163.895",
"value": "Hosokawa Yoriyuki"
},
{
"operation": "Highlighting",
"time": "173.698",
"value": "Hosokawa Yoriyuki_sentence_9"
},
{
"operation": "hyperlink",
"time": "175.568",
"value": "Ashikaga Yoshimitsu"
},
{
"operation": "Now on",
"time": "176.539",
"value": "Ashikaga Yoshimitsu"
},
{
"operation": "Page search",
"time": "185.194",
"value": "historian"
},
{
"operation": "finish",
"time": "210.17",
"value": "finish"
}
]
}
|
exfever_8644
|
Ellen Burstyn earned a nomination for Best Actress at an Academy Award for her appearance in the 1963 American supernatural horror film also starring a Swedish-French actor who appeared in more than 150 films and several television series in many languages.
|
{
"evidence": [
"Ellen Burstyn",
"The Exorcist (film)",
"Max von Sydow"
],
"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:\nEllen Burstyn earned a nomination for Best Actress at an Academy Award for her appearance in the 1963 American supernatural horror film also starring a Swedish-French actor who appeared in more than 150 films and several television series in many languages.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Ellen Burstyn",
"The Exorcist (film)",
"Max von Sydow"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2507,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
80090
|
Richard Vissing was mayor of Jeffersonville, Indiana, and was co-owner of the Vissings Dodge, established in 1968, and ran a wrecker service where he pulled in many cars during his days, with the most famous being that of John Dillinger.
|
{
"evidence": [
"Richard Vissing_sentence_5"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nRichard Vissing was mayor of Jeffersonville, Indiana, and was co-owner of the Vissings Dodge, established in 1968, and ran a wrecker service where he pulled in many cars during his days, with the most famous being that of John Dillinger.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Richard Vissing_sentence_5"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2508,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.74",
"value": "?search="
},
{
"operation": "search",
"time": "28.096",
"value": "Contains...Richard Vissing"
},
{
"operation": "Now on",
"time": "29.128",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "31.583",
"value": "Richard Vissing"
},
{
"operation": "Now on",
"time": "32.029",
"value": "Richard Vissing"
},
{
"operation": "Highlighting",
"time": "40.504",
"value": "Richard Vissing_sentence_0"
},
{
"operation": "Page search",
"time": "129.935",
"value": "dodge"
},
{
"operation": "Highlighting",
"time": "153.836",
"value": "Richard Vissing_sentence_8"
},
{
"operation": "Highlighting",
"time": "158.467",
"value": "Richard Vissing_sentence_5"
},
{
"operation": "Highlighting deleted",
"time": "168.553",
"value": "Richard Vissing_sentence_8"
},
{
"operation": "Highlighting deleted",
"time": "171.917",
"value": "Richard Vissing_sentence_0"
},
{
"operation": "finish",
"time": "316.447",
"value": "finish"
}
]
}
|
87159
|
Parlee Beach Provincial Park, a provincial park located in Pointe-du-Chêne, New Brunswick, Canada, was named in honor of T. Babbitt Parlee wherein it was established in 1957 with an area of 0.64 hectares while visitors for the park has been forbidden since 2010.
|
{
"evidence": [
"Parlee Beach Provincial Park_sentence_12"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nParlee Beach Provincial Park, a provincial park located in Pointe-du-Chêne, New Brunswick, Canada, was named in honor of T. Babbitt Parlee wherein it was established in 1957 with an area of 0.64 hectares while visitors for the park has been forbidden since 2010.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Parlee Beach Provincial Park_sentence_12"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2509,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.091",
"value": "?search="
},
{
"operation": "search",
"time": "11.947",
"value": "Parlee Beach Provincial Park"
},
{
"operation": "Now on",
"time": "12.69",
"value": "Parlee Beach Provincial Park"
},
{
"operation": "Highlighting",
"time": "28.504",
"value": "Parlee Beach Provincial Park_sentence_0"
},
{
"operation": "Highlighting",
"time": "41.888",
"value": "Parlee Beach Provincial Park_sentence_2Parlee Beach Provincial Park_sentence_3"
},
{
"operation": "Highlighting",
"time": "41.891",
"value": "Parlee Beach Provincial Park_section_1"
},
{
"operation": "Highlighting",
"time": "45.124",
"value": "Parlee Beach Provincial Park_cell_0_5_1"
},
{
"operation": "Highlighting",
"time": "45.132",
"value": "Parlee Beach Provincial Park_header_cell_0_5_0"
},
{
"operation": "Highlighting",
"time": "45.132",
"value": "Parlee Beach Provincial Park_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "52.686",
"value": "Parlee Beach Provincial Park_cell_0_4_1"
},
{
"operation": "Highlighting",
"time": "52.693",
"value": "Parlee Beach Provincial Park_header_cell_0_4_0"
},
{
"operation": "Highlighting",
"time": "79.845",
"value": "Parlee Beach Provincial Park_sentence_12"
},
{
"operation": "Highlighting deleted",
"time": "91.071",
"value": "Parlee Beach Provincial Park_sentence_2Parlee Beach Provincial Park_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "91.075",
"value": "Parlee Beach Provincial Park_section_1"
},
{
"operation": "Highlighting deleted",
"time": "93.248",
"value": "Parlee Beach Provincial Park_cell_0_5_1"
},
{
"operation": "Highlighting deleted",
"time": "93.256",
"value": "Parlee Beach Provincial Park_header_cell_0_5_0"
},
{
"operation": "Highlighting deleted",
"time": "93.986",
"value": "Parlee Beach Provincial Park_cell_0_4_1"
},
{
"operation": "Highlighting deleted",
"time": "93.994",
"value": "Parlee Beach Provincial Park_header_cell_0_4_0"
},
{
"operation": "Highlighting deleted",
"time": "93.996",
"value": "Parlee Beach Provincial Park_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "96.171",
"value": "Parlee Beach Provincial Park_sentence_0"
},
{
"operation": "finish",
"time": "121.039",
"value": "finish"
}
]
}
|
exfever_6176
|
Friends with Benefits is a 2011 film that starred a Canadian actor known for his role as Ben Epstein in How to Make It in America.
|
{
"evidence": [
"Friends with Benefits (film)",
"Bryan Greenberg"
],
"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:\nFriends with Benefits is a 2011 film that starred a Canadian actor known for his role as Ben Epstein in How to Make It in America.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Friends with Benefits (film)",
"Bryan Greenberg"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2510,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
78347
|
Riblja Čorba leader, Bora Đorđević, not only started performing and composing as a teenager; but also sang and played rhythm guitar in his early bands including Poslednji Ostatak Romantičnog Sveta.
|
{
"evidence": [
"Bora Đorđević_sentence_2",
"Bora Đorđević_sentence_5",
"Bora Đorđević_sentence_6",
"Bora Đorđević_sentence_8",
"Bora Đorđević_sentence_11",
"Bora Đorđević_sentence_12"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nRiblja Čorba leader, Bora Đorđević, not only started performing and composing as a teenager; but also sang and played rhythm guitar in his early bands including Poslednji Ostatak Romantičnog Sveta.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Bora Đorđević_sentence_2",
"Bora Đorđević_sentence_5",
"Bora Đorđević_sentence_6",
"Bora Đorđević_sentence_8",
"Bora Đorđević_sentence_11",
"Bora Đorđević_sentence_12"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2511,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.88",
"value": "?search="
},
{
"operation": "search",
"time": "446.829",
"value": "Bora Đorđević"
},
{
"operation": "Now on",
"time": "448.285",
"value": "Bora Đorđević"
},
{
"operation": "Highlighting",
"time": "458.16",
"value": "Bora Đorđević_sentence_2"
},
{
"operation": "Highlighting",
"time": "556.209",
"value": "Bora Đorđević_sentence_5"
},
{
"operation": "Highlighting",
"time": "570.706",
"value": "Bora Đorđević_sentence_8"
},
{
"operation": "Highlighting",
"time": "600.116",
"value": "Bora Đorđević_sentence_6"
},
{
"operation": "Highlighting deleted",
"time": "601.087",
"value": "Bora Đorđević_sentence_6"
},
{
"operation": "Highlighting deleted",
"time": "602.391",
"value": "Bora Đorđević_sentence_8"
},
{
"operation": "Highlighting",
"time": "604.14",
"value": "Bora Đorđević_sentence_8"
},
{
"operation": "search",
"time": "651.425",
"value": "Contains...Poslednji Ostatak Romantičnog Sveta"
},
{
"operation": "Now on",
"time": "652.613",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "656.606",
"value": "Bora Đorđević"
},
{
"operation": "Now on",
"time": "656.91",
"value": "Bora Đorđević"
},
{
"operation": "Highlighting",
"time": "668.123",
"value": "Bora Đorđević_sentence_11"
},
{
"operation": "Highlighting",
"time": "671.851",
"value": "Bora Đorđević_sentence_12"
},
{
"operation": "Highlighting deleted",
"time": "687.581",
"value": "Bora Đorđević_sentence_12"
},
{
"operation": "Highlighting deleted",
"time": "688.101",
"value": "Bora Đorđević_sentence_11"
},
{
"operation": "Highlighting deleted",
"time": "688.59",
"value": "Bora Đorđević_sentence_8"
},
{
"operation": "Highlighting",
"time": "689.213",
"value": "Bora Đorđević_sentence_6"
},
{
"operation": "Highlighting",
"time": "692.01",
"value": "Bora Đorđević_sentence_8"
},
{
"operation": "Highlighting",
"time": "697.501",
"value": "Bora Đorđević_sentence_11"
},
{
"operation": "Highlighting",
"time": "698.137",
"value": "Bora Đorđević_sentence_12"
},
{
"operation": "finish",
"time": "729.803",
"value": "finish"
}
]
}
|
82095
|
Eva Krížiková (who entered the film industry through a minor, backup role in Paľo Bielik's work The Mountains Are Stirring from 1952) had a voice acting role in the Cat on a Hot Tin Roof in 2005.
|
{
"evidence": [
"Eva Krížiková_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:\nEva Krížiková (who entered the film industry through a minor, backup role in Paľo Bielik's work The Mountains Are Stirring from 1952) had a voice acting role in the Cat on a Hot Tin Roof in 2005.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Eva Krížiková_sentence_4"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2512,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.312",
"value": "?search="
},
{
"operation": "search",
"time": "68.077",
"value": "Rajkumar"
},
{
"operation": "Now on",
"time": "69.259",
"value": "Rajkumar"
},
{
"operation": "hyperlink",
"time": "99.79",
"value": "Rajkumar (actor)"
},
{
"operation": "Now on",
"time": "101.784",
"value": "Rajkumar (actor)"
},
{
"operation": "search",
"time": "150.349",
"value": "Contains...The Mountains Are Stirring"
},
{
"operation": "Now on",
"time": "151.608",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "159.288",
"value": "Eva Krížiková"
},
{
"operation": "Now on",
"time": "160.174",
"value": "Eva Krížiková"
},
{
"operation": "Highlighting",
"time": "168.538",
"value": "Eva Krížiková_sentence_4"
},
{
"operation": "search",
"time": "225.357",
"value": "Cat on a Hot Tin Roof"
},
{
"operation": "Now on",
"time": "226.727",
"value": "Cat on a Hot Tin Roof"
},
{
"operation": "back-button-clicked",
"time": "266.299",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "266.392",
"value": "Eva Krížiková"
},
{
"operation": "Highlighting deleted",
"time": "295.178",
"value": "Eva Krížiková_sentence_4"
},
{
"operation": "Highlighting",
"time": "296.934",
"value": "Eva Krížiková_cell_0_2_1"
},
{
"operation": "Highlighting",
"time": "296.946",
"value": "Eva Krížiková_header_cell_0_2_0"
},
{
"operation": "Highlighting",
"time": "296.946",
"value": "Eva Krížiková_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "312.562",
"value": "Eva Krížiková_cell_0_2_1"
},
{
"operation": "Highlighting deleted",
"time": "312.576",
"value": "Eva Krížiková_header_cell_0_2_0"
},
{
"operation": "Highlighting deleted",
"time": "312.58",
"value": "Eva Krížiková_header_cell_0_0_0"
},
{
"operation": "Highlighting",
"time": "318.764",
"value": "Eva Krížiková_sentence_4"
},
{
"operation": "finish",
"time": "338.864",
"value": "finish"
}
]
}
|
61350
|
In Frank McDougall's first season, he made himself the top scorer in the Scottish Premier Division and helped The Dons win the league championship however, his second season saw fewer goals but did include a four-goal haul, wherein the Four goals are scored by a single player in a match, in a 4–1 over Celtic in November 1985.
|
{
"evidence": [
"Frank McDougall_sentence_10"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nIn Frank McDougall's first season, he made himself the top scorer in the Scottish Premier Division and helped The Dons win the league championship however, his second season saw fewer goals but did include a four-goal haul, wherein the Four goals are scored by a single player in a match, in a 4–1 over Celtic in November 1985.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Frank McDougall_sentence_10"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2513,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.812",
"value": "?search="
},
{
"operation": "search",
"time": "14.099",
"value": "Frank McDougall"
},
{
"operation": "Now on",
"time": "14.925",
"value": "Frank McDougall"
},
{
"operation": "Highlighting",
"time": "202.095",
"value": "Frank McDougall_sentence_8"
},
{
"operation": "Highlighting",
"time": "225.204",
"value": "Frank McDougall_sentence_10"
},
{
"operation": "search",
"time": "255.256",
"value": "four-goal haul"
},
{
"operation": "Now on",
"time": "258.713",
"value": "undefined"
},
{
"operation": "search",
"time": "332.738",
"value": "four goal haul"
},
{
"operation": "Now on",
"time": "333.433",
"value": "undefined"
},
{
"operation": "search",
"time": "379.429",
"value": "Frank McDougall"
},
{
"operation": "Now on",
"time": "380.337",
"value": "Frank McDougall"
},
{
"operation": "Highlighting",
"time": "387.831",
"value": "Frank McDougall_sentence_10"
},
{
"operation": "finish",
"time": "401.653",
"value": "finish"
}
]
}
|
8593
|
Todd Pillion, an American dentist, business owner, and politician from the Commonwealth of Virginia who graduated from the Virginia Commonwealth University School of Dentistry, joined the Virginia Army National Guard, with the commission of captain.
|
{
"evidence": [
"Todd Pillion_sentence_0",
"Todd Pillion_sentence_5",
"Todd Pillion_sentence_6"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nTodd Pillion, an American dentist, business owner, and politician from the Commonwealth of Virginia who graduated from the Virginia Commonwealth University School of Dentistry, joined the Virginia Army National Guard, with the commission of captain.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Todd Pillion_sentence_0",
"Todd Pillion_sentence_5",
"Todd Pillion_sentence_6"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2514,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.72",
"value": "?search="
},
{
"operation": "search",
"time": "8.859",
"value": "Todd Pillion"
},
{
"operation": "Now on",
"time": "10.205",
"value": "Todd Pillion"
},
{
"operation": "Highlighting",
"time": "12.453",
"value": "Todd Pillion_sentence_0"
},
{
"operation": "Highlighting",
"time": "23.335",
"value": "Todd Pillion_sentence_5"
},
{
"operation": "Highlighting",
"time": "29.149",
"value": "Todd Pillion_sentence_6"
},
{
"operation": "finish",
"time": "69.012",
"value": "finish"
}
]
}
|
79207
|
The National Junior Art Honor Society (NJAHS) was established in the United States in 1989 by the National Art Education Association (NEAE) to neither help recognize nor inspire interested students to do more with their creative ability.
|
{
"evidence": [
"National Junior Art Honor Society_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 National Junior Art Honor Society (NJAHS) was established in the United States in 1989 by the National Art Education Association (NEAE) to neither help recognize nor inspire interested students to do more with their creative ability.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"National Junior Art Honor Society_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2515,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.034",
"value": "?search="
},
{
"operation": "search",
"time": "12.035",
"value": "National Junior Art Honor Society"
},
{
"operation": "Now on",
"time": "12.727",
"value": "National Junior Art Honor Society"
},
{
"operation": "Highlighting",
"time": "17.709",
"value": "National Junior Art Honor Society_sentence_0"
},
{
"operation": "Highlighting",
"time": "24.806",
"value": "National Junior Art Honor Society_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "30.942",
"value": "National Junior Art Honor Society_sentence_0"
},
{
"operation": "finish",
"time": "38.168",
"value": "finish"
}
]
}
|
exfever_589
|
Sage Moonblood Stallone was the wealthiest American actor.
|
{
"evidence": [
"Sage Stallone",
"Grindhouse Releasing"
],
"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:\nSage Moonblood Stallone was the wealthiest American actor.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Sage Stallone",
"Grindhouse Releasing"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2516,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6949
|
Henry V is a film directed by an English theatre director.
|
{
"evidence": [
"Henry V (2012 film)",
"Thea Sharrock"
],
"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 V is a film directed by an English theatre director.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Henry V (2012 film)",
"Thea Sharrock"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2517,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5001
|
A ministate is a sovereign state with a small population and land like an island country in the Pacific.
|
{
"evidence": [
"Ministate",
"Palau"
],
"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:\nA ministate is a sovereign state with a small population and land like an island country in the Pacific.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Ministate",
"Palau"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2518,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6562
|
Persuasion is a 2007 British television film adaptation of a novel by an English novelist who achieved modest success with the publication of her novels, including an 1813 novel of manners.
|
{
"evidence": [
"Persuasion (2007 film)",
"Jane Austen",
"Pride and Prejudice"
],
"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:\nPersuasion is a 2007 British television film adaptation of a novel by an English novelist who achieved modest success with the publication of her novels, including an 1813 novel of manners.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Persuasion (2007 film)",
"Jane Austen",
"Pride and Prejudice"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2519,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4272
|
Joseph Michael "Joey" Graceffa runs four active channels in Mexican online video sharing and social media platforms.
|
{
"evidence": [
"Joey Graceffa",
"YouTube"
],
"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:\nJoseph Michael \"Joey\" Graceffa runs four active channels in Mexican online video sharing and social media platforms.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Joey Graceffa",
"YouTube"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2520,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
77859
|
Curon is an Italian supernatural drama television series created by Ezio Abbate, Ivano Fachin, Giovanni Galassi, and Tommaso Matano for Netflix that starred Valeria Bilello (born 2 May 1982, Sciacca, Sicily, Italy) and Luca Lionello, and had 7 episodes that were all simultaneously released on June 10, 2019.
|
{
"evidence": [
"Curon_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:\nCuron is an Italian supernatural drama television series created by Ezio Abbate, Ivano Fachin, Giovanni Galassi, and Tommaso Matano for Netflix that starred Valeria Bilello (born 2 May 1982, Sciacca, Sicily, Italy) and Luca Lionello, and had 7 episodes that were all simultaneously released on June 10, 2019.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Curon_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2521,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.275",
"value": "?search="
},
{
"operation": "search",
"time": "7.864",
"value": "Curon"
},
{
"operation": "Now on",
"time": "8.89",
"value": "Curon"
},
{
"operation": "Highlighting",
"time": "18.324",
"value": "Curon_sentence_2"
},
{
"operation": "Highlighting",
"time": "31.74",
"value": "Curon_sentence_3"
},
{
"operation": "hyperlink",
"time": "34.551",
"value": "Valeria Bilello"
},
{
"operation": "Now on",
"time": "35.192",
"value": "Valeria Bilello"
},
{
"operation": "Highlighting",
"time": "37.461",
"value": "Valeria Bilello_sentence_0"
},
{
"operation": "Highlighting",
"time": "37.462",
"value": "Valeria Bilello_title"
},
{
"operation": "back-button-clicked",
"time": "40.176",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "40.24",
"value": "Curon"
},
{
"operation": "Highlighting",
"time": "51.107",
"value": "Curon_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "53.331",
"value": "Curon_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "53.887",
"value": "Curon_sentence_2"
},
{
"operation": "finish",
"time": "76.846",
"value": "finish"
}
]
}
|
48953
|
The United Nations Security Council Resolution 858 was adopted on August 24, 1993, at meeting number 3,268, after recalling resolutions 849 and 854 and noting a ceasefire between Abkhazia and the Philippines.
|
{
"evidence": [
"United Nations Security Council Resolution 858_sentence_0"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe United Nations Security Council Resolution 858 was adopted on August 24, 1993, at meeting number 3,268, after recalling resolutions 849 and 854 and noting a ceasefire between Abkhazia and the Philippines.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"United Nations Security Council Resolution 858_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2522,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.248",
"value": "?search="
},
{
"operation": "search",
"time": "16.666",
"value": "United Nations Security Council Resolution 858"
},
{
"operation": "Now on",
"time": "17.588",
"value": "United Nations Security Council Resolution 858"
},
{
"operation": "Highlighting",
"time": "27.556",
"value": "United Nations Security Council Resolution 858_sentence_0"
},
{
"operation": "Highlighting",
"time": "34.39",
"value": "United Nations Security Council Resolution 858_cell_0_2_1"
},
{
"operation": "Highlighting",
"time": "34.394",
"value": "United Nations Security Council Resolution 858_header_cell_0_2_0"
},
{
"operation": "Highlighting",
"time": "34.395",
"value": "United Nations Security Council Resolution 858_header_cell_0_0_0"
},
{
"operation": "Highlighting deleted",
"time": "61.925",
"value": "United Nations Security Council Resolution 858_cell_0_2_1"
},
{
"operation": "Highlighting deleted",
"time": "61.93",
"value": "United Nations Security Council Resolution 858_header_cell_0_2_0"
},
{
"operation": "Highlighting deleted",
"time": "61.932",
"value": "United Nations Security Council Resolution 858_header_cell_0_0_0"
},
{
"operation": "finish",
"time": "73.922",
"value": "finish"
}
]
}
|
73316
|
Michigan State Normal Normalites played against Olivet College in October 17 and won 28-3.
|
{
"evidence": [
"Eastern Michigan Eagles football_sentence_134"
],
"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:\nMichigan State Normal Normalites played against Olivet College in October 17 and won 28-3.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Eastern Michigan Eagles football_sentence_134"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2523,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.652",
"value": "?search="
},
{
"operation": "search",
"time": "16.247",
"value": "Michigan State Normal Normalites football"
},
{
"operation": "Now on",
"time": "21.281",
"value": "Eastern Michigan Eagles football"
},
{
"operation": "search",
"time": "45.725",
"value": "Michigan State Normal Normalites 2017"
},
{
"operation": "Now on",
"time": "46.874",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "55.244",
"value": "Eastern Michigan Eagles football"
},
{
"operation": "Now on",
"time": "60.765",
"value": "Eastern Michigan Eagles football"
},
{
"operation": "search",
"time": "176.008",
"value": "Michigan State Normal Normalites vs Olivet College"
},
{
"operation": "Now on",
"time": "176.709",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "182.412",
"value": "Eastern Michigan Eagles football"
},
{
"operation": "Now on",
"time": "183.015",
"value": "Eastern Michigan Eagles football"
},
{
"operation": "Page search",
"time": "188.771",
"value": "Olivet"
},
{
"operation": "Highlighting",
"time": "198.063",
"value": "Eastern Michigan Eagles football_sentence_134"
},
{
"operation": "Highlighting deleted",
"time": "199.996",
"value": "Eastern Michigan Eagles football_sentence_134"
},
{
"operation": "search",
"time": "321.403",
"value": "Normal Normalities"
},
{
"operation": "Now on",
"time": "322.638",
"value": "undefined"
},
{
"operation": "search",
"time": "340.366",
"value": "2017 Normal Normalities"
},
{
"operation": "Now on",
"time": "342.007",
"value": "undefined"
},
{
"operation": "search",
"time": "359.007",
"value": "Michigan State Normal Normalites football"
},
{
"operation": "Now on",
"time": "360.131",
"value": "Eastern Michigan Eagles football"
},
{
"operation": "search",
"time": "400.261",
"value": "Michigan State Normal Normalites men's basketball"
},
{
"operation": "Now on",
"time": "413.139",
"value": "Eastern Michigan Eagles men's basketball"
},
{
"operation": "Page search",
"time": "424.289",
"value": "Olivet"
},
{
"operation": "search",
"time": "443.529",
"value": "Michigan State Normal Normalites football 2017"
},
{
"operation": "Now on",
"time": "444.036",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "449.215",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "453.34",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "453.631",
"value": "Eastern Michigan Eagles football"
},
{
"operation": "search",
"time": "465.358",
"value": "Michigan State Normal Normalites football, 1926"
},
{
"operation": "Now on",
"time": "466.417",
"value": "1926 Michigan State Normal Normalites football team"
},
{
"operation": "search",
"time": "498.657",
"value": "Michigan State Normal Normalites football October 17"
},
{
"operation": "Now on",
"time": "500.126",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "506.868",
"value": "1911 Michigan State Normal Normalites football team"
},
{
"operation": "Now on",
"time": "507.413",
"value": "1911 Michigan State Normal Normalites football team"
},
{
"operation": "search",
"time": "516.486",
"value": "Michigan State Normal Normalites football, 1922"
},
{
"operation": "Now on",
"time": "517.326",
"value": "1922 Michigan State Normal Normalites football team"
},
{
"operation": "search",
"time": "523.285",
"value": "Michigan State Normal Normalites football, 1921"
},
{
"operation": "Now on",
"time": "524.108",
"value": "1921 Michigan State Normal Normalites football team"
},
{
"operation": "search",
"time": "529.652",
"value": "Michigan State Normal Normalites football, 1923"
},
{
"operation": "Now on",
"time": "530.552",
"value": "1923 Michigan State Normal Normalites football team"
},
{
"operation": "search",
"time": "545.647",
"value": "Michigan State Normal Normalites football, 1904"
},
{
"operation": "Now on",
"time": "546.765",
"value": "Michigan State Normal Normalites football, 1904"
},
{
"operation": "search",
"time": "577.526",
"value": "Michigan State Normal Normalites football, 1923"
},
{
"operation": "Now on",
"time": "578.167",
"value": "Michigan State Normal Normalites football, 1923"
},
{
"operation": "search",
"time": "594.11",
"value": "Michigan State Normal Normalites football, 1926"
},
{
"operation": "Now on",
"time": "594.9",
"value": "Michigan State Normal Normalites football, 1926"
},
{
"operation": "hyperlink",
"time": "604.542",
"value": "Michigan State Normal Normalites football, 1926"
},
{
"operation": "Now on",
"time": "605.404",
"value": "undefined"
},
{
"operation": "search",
"time": "609.463",
"value": "Michigan State Normal Normalites football"
},
{
"operation": "Now on",
"time": "610.227",
"value": "Michigan State Normal Normalites football"
},
{
"operation": "hyperlink",
"time": "657.077",
"value": "Liberty"
},
{
"operation": "Now on",
"time": "661.986",
"value": "Liberty Flames football"
},
{
"operation": "search",
"time": "678.045",
"value": "Michigan State Normal Normalites 28-3"
},
{
"operation": "Now on",
"time": "679.951",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "697.372",
"value": "1904 Michigan State Normal Normalites football team"
},
{
"operation": "Now on",
"time": "698.036",
"value": "1904 Michigan State Normal Normalites football team"
},
{
"operation": "search",
"time": "743.095",
"value": "Michigan State Normal Normalites football"
},
{
"operation": "Now on",
"time": "744.932",
"value": "Eastern Michigan Eagles football"
},
{
"operation": "Page search",
"time": "755.994",
"value": "Olivet"
},
{
"operation": "search",
"time": "783.477",
"value": "Michigan State Normal Normalites 1906"
},
{
"operation": "Now on",
"time": "784.353",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "786.084",
"value": "1906 Michigan State Normal Normalites football team"
},
{
"operation": "Now on",
"time": "786.809",
"value": "1906 Michigan State Normal Normalites football team"
},
{
"operation": "search",
"time": "818.727",
"value": "Michigan State Normal Normalites football"
},
{
"operation": "Now on",
"time": "819.673",
"value": "Michigan State Normal Normalites football"
},
{
"operation": "Page search",
"time": "822.16",
"value": "Olivet"
},
{
"operation": "Highlighting",
"time": "839.637",
"value": "Eastern Michigan Eagles football_sentence_133"
},
{
"operation": "Highlighting deleted",
"time": "841.62",
"value": "Eastern Michigan Eagles football_sentence_133"
},
{
"operation": "search",
"time": "867.127",
"value": "Michigan Intercollegiate Athletic Association conference champinoships"
},
{
"operation": "Now on",
"time": "869.746",
"value": "undefined"
},
{
"operation": "search",
"time": "878.891",
"value": "Michigan Intercollegiate Athletic Association"
},
{
"operation": "Now on",
"time": "882.798",
"value": "Michigan Intercollegiate Athletic Association"
},
{
"operation": "search",
"time": "906.158",
"value": "Michigan State Normal Normalites football"
},
{
"operation": "Now on",
"time": "907.083",
"value": "Michigan State Normal Normalites football"
},
{
"operation": "Page search",
"time": "912.259",
"value": "olivet"
},
{
"operation": "Highlighting",
"time": "920.99",
"value": "Eastern Michigan Eagles football_sentence_134"
},
{
"operation": "search",
"time": "983.728",
"value": "Eastern Michigan Yearly results"
},
{
"operation": "Now on",
"time": "985.896",
"value": "undefined"
},
{
"operation": "search",
"time": "1041.139",
"value": "eastern michigan football conference"
},
{
"operation": "Now on",
"time": "1042.637",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "1047.164",
"value": "Mid-American Conference"
},
{
"operation": "Now on",
"time": "1049.913",
"value": "Mid-American Conference"
},
{
"operation": "search",
"time": "1204.566",
"value": "Olivet"
},
{
"operation": "Now on",
"time": "1205.383",
"value": "Olivet"
},
{
"operation": "search",
"time": "1222.687",
"value": "Michigan State Normal Normalites football"
},
{
"operation": "Now on",
"time": "1223.605",
"value": "Michigan State Normal Normalites football"
},
{
"operation": "Page search",
"time": "1250.551",
"value": "olivet"
},
{
"operation": "search",
"time": "1310.308",
"value": "Mid-American Conference Championship"
},
{
"operation": "Now on",
"time": "1311.093",
"value": "Mid-American Conference Championship"
},
{
"operation": "hyperlink",
"time": "1316.587",
"value": "MAC Championship Game"
},
{
"operation": "Now on",
"time": "1318.39",
"value": "MAC Football Championship Game"
},
{
"operation": "search",
"time": "1341.313",
"value": "Normal Normalites vs Olivet"
},
{
"operation": "Now on",
"time": "1342.672",
"value": "undefined"
},
{
"operation": "search",
"time": "1350.273",
"value": "Olivet football"
},
{
"operation": "Now on",
"time": "1351.216",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "1355.628",
"value": "1910 Olivet football team"
},
{
"operation": "Now on",
"time": "1356.171",
"value": "1910 Olivet football team"
},
{
"operation": "back-button-clicked",
"time": "1361.621",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "1362.363",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "1364.454",
"value": "1909 Olivet football team"
},
{
"operation": "Now on",
"time": "1364.971",
"value": "1909 Olivet football team"
},
{
"operation": "back-button-clicked",
"time": "1367.069",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "1367.859",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "1369.685",
"value": "1907 Olivet football team"
},
{
"operation": "Now on",
"time": "1370.357",
"value": "1907 Olivet football team"
},
{
"operation": "back-button-clicked",
"time": "1375.924",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "1376.654",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "1378.661",
"value": "1894 Olivet football team"
},
{
"operation": "Now on",
"time": "1379.304",
"value": "1894 Olivet football team"
},
{
"operation": "back-button-clicked",
"time": "1393.173",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "1393.935",
"value": "undefined"
},
{
"operation": "finish",
"time": "1424.372",
"value": "finish"
}
]
}
|
60834
|
Claude Champion de Crespigny received the substantive rank of Captain in his regiment on January 12, 1902; then later became the Aide-de-Camp to the Viceroy and Governor-General of India George Curzon, 1st Marquess Curzon of Kedleston (styled as Lord Curzon of Kedleston between 1898 and 1911).
|
{
"evidence": [
"Claude Champion de Crespigny_sentence_5",
"Claude Champion de Crespigny_sentence_6",
"George Curzon, 1st Marquess Curzon of Kedleston_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:\nClaude Champion de Crespigny received the substantive rank of Captain in his regiment on January 12, 1902; then later became the Aide-de-Camp to the Viceroy and Governor-General of India George Curzon, 1st Marquess Curzon of Kedleston (styled as Lord Curzon of Kedleston between 1898 and 1911).\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Claude Champion de Crespigny_sentence_5",
"Claude Champion de Crespigny_sentence_6",
"George Curzon, 1st Marquess Curzon of Kedleston_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2524,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.182",
"value": "?search="
},
{
"operation": "search",
"time": "4.54",
"value": "Claude Champion de Crespigny"
},
{
"operation": "Now on",
"time": "5.295",
"value": "Claude Champion de Crespigny"
},
{
"operation": "Highlighting",
"time": "26.404",
"value": "Claude Champion de Crespigny_sentence_5"
},
{
"operation": "Highlighting",
"time": "32.182",
"value": "Claude Champion de Crespigny_sentence_6"
},
{
"operation": "hyperlink",
"time": "38.721",
"value": "George Curzon, 1st Marquess Curzon of Kedleston"
},
{
"operation": "Now on",
"time": "40.459",
"value": "George Curzon, 1st Marquess Curzon of Kedleston"
},
{
"operation": "Highlighting",
"time": "53.199",
"value": "George Curzon, 1st Marquess Curzon of Kedleston_sentence_0"
},
{
"operation": "finish",
"time": "61.076",
"value": "finish"
}
]
}
|
14140
|
Tina Turner has written a total of more than 30 poems from 1969 to 1977.
|
{
"evidence": [
"Tina Turner_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:\nTina Turner has written a total of more than 30 poems from 1969 to 1977.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Tina Turner_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2525,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0",
"value": "?search="
},
{
"operation": "search",
"time": "0.392",
"value": "tina turner"
},
{
"operation": "Now on",
"time": "0.394",
"value": "Tina turner"
},
{
"operation": "Highlighting",
"time": "0.51",
"value": "Tina Turner_sentence_0"
},
{
"operation": "finish",
"time": "0.558",
"value": "finish"
},
{
"operation": "Now on",
"time": "1619479053.323",
"value": "Tina Turner"
},
{
"operation": "finish",
"time": "1619479057.262",
"value": "finish"
}
]
}
|
exfever_7725
|
Mariah Carey is one of the best singers in America.
|
{
"evidence": [
"Touch My Body",
"The-Dream"
],
"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:\nMariah Carey is one of the best singers in America.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Touch My Body",
"The-Dream"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2526,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1760
|
Carole King is an American singer who experienced commercial disappointment with her debut studio album, which included a song co-written by Gerry Goffin and recorded by The Drifters in 1962.
|
{
"evidence": [
"Carole King",
"Writer (album)",
"Up on the Roof (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:\nCarole King is an American singer who experienced commercial disappointment with her debut studio album, which included a song co-written by Gerry Goffin and recorded by The Drifters in 1962.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Carole King",
"Writer (album)",
"Up on the Roof (song)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2527,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8402
|
The Lord of the Rings series is the highest-selling novel book of all time.
|
{
"evidence": [
"George R. R. Martin",
"Short story"
],
"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 Lord of the Rings series is the highest-selling novel book of all time.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"George R. R. Martin",
"Short story"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2528,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1033
|
Debut studio album by an American singer who was born on December 14, 1992.
|
{
"evidence": [
"Tori Kelly",
"Unbreakable Smile",
"Tori Kelly"
],
"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:\nDebut studio album by an American singer who was born on December 14, 1992.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Tori Kelly",
"Unbreakable Smile",
"Tori Kelly"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2529,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
47439
|
Stoebe is a genus of animals.
|
{
"evidence": [
"Stoebe_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:\nStoebe is a genus of animals.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Stoebe_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2530,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.456",
"value": "?search="
},
{
"operation": "search",
"time": "8.566",
"value": "Stoebe"
},
{
"operation": "Now on",
"time": "9.488",
"value": "Stoebe"
},
{
"operation": "Highlighting",
"time": "12.87",
"value": "Stoebe_sentence_0"
},
{
"operation": "Now on",
"time": "184.209",
"value": "Stoebe"
},
{
"operation": "finish",
"time": "244.326",
"value": "finish"
}
]
}
|
exfever_2913
|
The Aix-Marseille-Provence Metropolis is a métropole, an intercommunal structure, centered on the cities, including a city and a commune in western France.
|
{
"evidence": [
"Aix-Marseille-Provence Metropolis",
"Aix-en-Provence"
],
"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 Aix-Marseille-Provence Metropolis is a métropole, an intercommunal structure, centered on the cities, including a city and a commune in western France.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Aix-Marseille-Provence Metropolis",
"Aix-en-Provence"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2531,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
25251
|
Dogboy recorded the debut Too Rude which was released with Suburban Noize Records on May 4, 2000; then after the first album, the second Re-Invention was not released until March 26, 2004.
|
{
"evidence": [
"Dog Boy_sentence_4",
"Dog Boy_sentence_5"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nDogboy recorded the debut Too Rude which was released with Suburban Noize Records on May 4, 2000; then after the first album, the second Re-Invention was not released until March 26, 2004.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Dog Boy_sentence_4",
"Dog Boy_sentence_5"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2532,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.496",
"value": "?search="
},
{
"operation": "search",
"time": "20.12",
"value": "Too Rude (album)"
},
{
"operation": "Now on",
"time": "20.813",
"value": "Too Rude (album)"
},
{
"operation": "Highlighting",
"time": "48.91",
"value": "Too Rude (album)_sentence_1"
},
{
"operation": "Highlighting",
"time": "48.91",
"value": "Too Rude (album)_title"
},
{
"operation": "Highlighting",
"time": "54.317",
"value": "Too Rude (album)_sentence_2"
},
{
"operation": "hyperlink",
"time": "63.093",
"value": "Dogboy"
},
{
"operation": "Now on",
"time": "63.706",
"value": "Dogboy"
},
{
"operation": "Highlighting",
"time": "85.436",
"value": "Dog Boy_sentence_4"
},
{
"operation": "Highlighting",
"time": "110.332",
"value": "Dog Boy_sentence_5"
},
{
"operation": "finish",
"time": "118.908",
"value": "finish"
}
]
}
|
9272
|
In 1850, the Lungkwitz and Petri families emigrated to the United States, landing first in New York City, then migrated to Wheeling, West Virginia, but ultimately ending up in the settlement of Pedernales, Texas near Fredericksburg in 1851.
|
{
"evidence": [
"Friedrich Richard Petri_sentence_10"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nIn 1850, the Lungkwitz and Petri families emigrated to the United States, landing first in New York City, then migrated to Wheeling, West Virginia, but ultimately ending up in the settlement of Pedernales, Texas near Fredericksburg in 1851.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Friedrich Richard Petri_sentence_10"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2533,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.084",
"value": "?search="
},
{
"operation": "search",
"time": "55.318",
"value": "lungkwitz family"
},
{
"operation": "Now on",
"time": "55.859",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "60.506",
"value": "Friedrich Richard Petri"
},
{
"operation": "Now on",
"time": "61.366",
"value": "Friedrich Richard Petri"
},
{
"operation": "Highlighting",
"time": "77.905",
"value": "Friedrich Richard Petri_sentence_7"
},
{
"operation": "Highlighting",
"time": "82.403",
"value": "Friedrich Richard Petri_sentence_8"
},
{
"operation": "Highlighting",
"time": "117.29",
"value": "Friedrich Richard Petri_sentence_10"
},
{
"operation": "Highlighting deleted",
"time": "118.952",
"value": "Friedrich Richard Petri_sentence_8"
},
{
"operation": "Highlighting deleted",
"time": "119.794",
"value": "Friedrich Richard Petri_sentence_7"
},
{
"operation": "finish",
"time": "134.237",
"value": "finish"
}
]
}
|
exfever_6414
|
Marvel Studios began independent production on 2019.
|
{
"evidence": [
"Marvel Cinematic Universe",
"American comic book"
],
"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:\nMarvel Studios began independent production on 2019.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Marvel Cinematic Universe",
"American comic book"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2534,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6737
|
Views is a studio album released by an African record label imprint of Cash Money Records and distributed by Republic Records.
|
{
"evidence": [
"Views (album)",
"Young Money Entertainment"
],
"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:\nViews is a studio album released by an African record label imprint of Cash Money Records and distributed by Republic Records.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Views (album)",
"Young Money Entertainment"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2535,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7641
|
Ruth Negga is known for the television series Preacher, which was released on an American multinational basic cable television channel.
|
{
"evidence": [
"Ruth Negga",
"AMC (TV channel)"
],
"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:\nRuth Negga is known for the television series Preacher, which was released on an American multinational basic cable television channel.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Ruth Negga",
"AMC (TV channel)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2536,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5259
|
An opera is a collaborative form of performing art that uses live performers, usually dancers or actresses, to present the experience of a real or imagined event before a live audience in a specific place, often a stage.
|
{
"evidence": [
"Opera",
"Theatre"
],
"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:\nAn opera is a collaborative form of performing art that uses live performers, usually dancers or actresses, to present the experience of a real or imagined event before a live audience in a specific place, often a stage.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Opera",
"Theatre"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2537,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2319
|
The Black Eyed Peas were originally a group whose genre is a subgenre of hip hop music that encompasses the wide range of styles that are not typically identified as mainstream.
|
{
"evidence": [
"The Black Eyed Peas",
"Alternative hip hop"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Black Eyed Peas were originally a group whose genre is a subgenre of hip hop music that encompasses the wide range of styles that are not typically identified as mainstream.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Black Eyed Peas",
"Alternative hip hop"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2538,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
23374
|
Rochford, an Irish hurler who played as a full-back for the Kilkenny senior, added a second Leinster medal to his collection in 1904; then the subsequent All-Ireland final was delayed until June 24,1906.
|
{
"evidence": [
"Jack Rochford_sentence_0",
"Jack Rochford_sentence_25",
"Jack Rochford_sentence_26"
],
"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:\nRochford, an Irish hurler who played as a full-back for the Kilkenny senior, added a second Leinster medal to his collection in 1904; then the subsequent All-Ireland final was delayed until June 24,1906.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Jack Rochford_sentence_0",
"Jack Rochford_sentence_25",
"Jack Rochford_sentence_26"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2539,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.751",
"value": "?search="
},
{
"operation": "search",
"time": "14.137",
"value": "Rochford"
},
{
"operation": "Now on",
"time": "15.091",
"value": "Rochford"
},
{
"operation": "hyperlink",
"time": "18.208",
"value": "Rochford (disambiguation)"
},
{
"operation": "Now on",
"time": "19.078",
"value": "Rochford (disambiguation)"
},
{
"operation": "hyperlink",
"time": "25.939",
"value": "Rochford (surname)"
},
{
"operation": "Now on",
"time": "26.553",
"value": "Rochford (surname)"
},
{
"operation": "hyperlink",
"time": "32.195",
"value": "Bernard Rochford"
},
{
"operation": "Now on",
"time": "33.096",
"value": "Bernard Rochford"
},
{
"operation": "Highlighting",
"time": "36.856",
"value": "Bernard Rochford_sentence_0"
},
{
"operation": "Highlighting",
"time": "36.857",
"value": "Bernard Rochford_title"
},
{
"operation": "Page search",
"time": "59.124",
"value": "kilkenny"
},
{
"operation": "Page search",
"time": "107.663",
"value": "leinster"
},
{
"operation": "search",
"time": "116.516",
"value": "Contains...leinster medal"
},
{
"operation": "Now on",
"time": "117.917",
"value": "undefined"
},
{
"operation": "Page search",
"time": "145.625",
"value": "\"added a second leinster medal\""
},
{
"operation": "Page search",
"time": "162.962",
"value": "\"added a second leinster medal\""
},
{
"operation": "search",
"time": "181.527",
"value": "\"added a second leinster medal\""
},
{
"operation": "Now on",
"time": "182.91",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "214.826",
"value": "Jack Rochford"
},
{
"operation": "Now on",
"time": "216.332",
"value": "Jack Rochford"
},
{
"operation": "Highlighting",
"time": "219.818",
"value": "Jack Rochford_sentence_0"
},
{
"operation": "Page search",
"time": "233.611",
"value": "leinster medal"
},
{
"operation": "Highlighting",
"time": "242.593",
"value": "Jack Rochford_sentence_25"
},
{
"operation": "Highlighting",
"time": "247.012",
"value": "Jack Rochford_sentence_26"
},
{
"operation": "finish",
"time": "268.989",
"value": "finish"
}
]
}
|
70065
|
Catherine Ibarguen scored twelve points for the first time in the Diamond race, which put her in a tie with one other competitor.
|
{
"evidence": [
"Caterine Ibargüen_sentence_10"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nCatherine Ibarguen scored twelve points for the first time in the Diamond race, which put her in a tie with one other competitor.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Caterine Ibargüen_sentence_10"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2540,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "49.714",
"value": "undefined"
},
{
"operation": "search",
"time": "55.003",
"value": "Catherine Ibarguen"
},
{
"operation": "Now on",
"time": "57.372",
"value": "Catherine Ibarguen"
},
{
"operation": "Page search",
"time": "92.687",
"value": "tie"
},
{
"operation": "Page search",
"time": "93.9",
"value": "tie"
},
{
"operation": "Page search",
"time": "120.451",
"value": "points"
},
{
"operation": "hyperlink",
"time": "136.197",
"value": "armed conflict in Colombia"
},
{
"operation": "Now on",
"time": "138.668",
"value": "Colombian conflict (1964–present)"
},
{
"operation": "Page search",
"time": "149.995",
"value": "Catherine"
},
{
"operation": "back-button-clicked",
"time": "162.255",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "163.605",
"value": "Caterine Ibargüen"
},
{
"operation": "Page search",
"time": "185.012",
"value": "Diamond"
},
{
"operation": "Highlighting",
"time": "221.698",
"value": "Caterine Ibargüen_sentence_10"
},
{
"operation": "hyperlink",
"time": "227.215",
"value": "Diamond League"
},
{
"operation": "Now on",
"time": "233.895",
"value": "Caterine Ibargüen"
},
{
"operation": "finish",
"time": "242.945",
"value": "finish"
}
]
}
|
exfever_6009
|
Snow White and the Seven Dwarfs is the highest-grossing film in 1937.
|
{
"evidence": [
"Evil Queen (Disney)",
"Walt Disney Animation Studios"
],
"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:\nSnow White and the Seven Dwarfs is the highest-grossing film in 1937.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Evil Queen (Disney)",
"Walt Disney Animation Studios"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2541,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
43096
|
Arabian Peninsula's tropical cyclones are spawned in the Arabian Sea and the Indian Ocean, which are being tracked by the India Meteorological Department and the American-based Joint Typhoon Warning Center.
|
{
"evidence": [
"List of Arabian Peninsula tropical cyclones_sentence_3",
"List of Arabian Peninsula tropical cyclones_sentence_8",
"List of Arabian Peninsula tropical cyclones_sentence_9"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nArabian Peninsula's tropical cyclones are spawned in the Arabian Sea and the Indian Ocean, which are being tracked by the India Meteorological Department and the American-based Joint Typhoon Warning Center.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"List of Arabian Peninsula tropical cyclones_sentence_3",
"List of Arabian Peninsula tropical cyclones_sentence_8",
"List of Arabian Peninsula tropical cyclones_sentence_9"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2542,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.138",
"value": "?search="
},
{
"operation": "search",
"time": "11.822",
"value": "Arabian Peninsula"
},
{
"operation": "Now on",
"time": "12.983",
"value": "Arabian Peninsula"
},
{
"operation": "search",
"time": "55.213",
"value": "Arabian Peninsula's tropical cyclones"
},
{
"operation": "Now on",
"time": "56.381",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "62.02",
"value": "North Indian Ocean tropical cyclone"
},
{
"operation": "Now on",
"time": "62.817",
"value": "North Indian Ocean tropical cyclone"
},
{
"operation": "search",
"time": "106.697",
"value": "India Meteorological Department"
},
{
"operation": "Now on",
"time": "107.406",
"value": "India Meteorological Department"
},
{
"operation": "hyperlink",
"time": "141.692",
"value": "tropical cyclones"
},
{
"operation": "Now on",
"time": "142.911",
"value": "Tropical cyclone"
},
{
"operation": "search",
"time": "170.549",
"value": "Contains...tracked by the India Meteorological Department"
},
{
"operation": "Now on",
"time": "172.009",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "188.653",
"value": "List of Arabian Peninsula tropical cyclones"
},
{
"operation": "Now on",
"time": "189.41",
"value": "List of Arabian Peninsula tropical cyclones"
},
{
"operation": "Highlighting",
"time": "200.566",
"value": "List of Arabian Peninsula tropical cyclones_sentence_3"
},
{
"operation": "Highlighting",
"time": "236.132",
"value": "List of Arabian Peninsula tropical cyclones_sentence_8"
},
{
"operation": "Highlighting",
"time": "240.628",
"value": "List of Arabian Peninsula tropical cyclones_sentence_9"
},
{
"operation": "Highlighting",
"time": "252.58",
"value": "List of Arabian Peninsula tropical cyclones_sentence_8"
},
{
"operation": "Highlighting",
"time": "252.964",
"value": "List of Arabian Peninsula tropical cyclones_sentence_9"
},
{
"operation": "Highlighting",
"time": "253.404",
"value": "List of Arabian Peninsula tropical cyclones_sentence_10"
},
{
"operation": "Highlighting",
"time": "285.045",
"value": "List of Arabian Peninsula tropical cyclones_sentence_3"
},
{
"operation": "finish",
"time": "331.768",
"value": "finish"
}
]
}
|
exfever_6291
|
Richard St John Francis Harris was the highest-paid Irish actor in 2000s.
|
{
"evidence": [
"Richard Harris",
"Oliver Cromwell",
"Charles I of England"
],
"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:\nRichard St John Francis Harris was the highest-paid Irish actor in 2000s.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Richard Harris",
"Oliver Cromwell",
"Charles I of England"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2543,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
19060
|
Onetangi Sports Park-based Waiheke United AFC adopted the Whole of Football Plan, which provides a comprehensive model with coherent pathways and age-appropriate skills development.
|
{
"evidence": [
"Waiheke United AFC_sentence_0",
"Waiheke United AFC_sentence_4",
"Waiheke United AFC_sentence_6"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nOnetangi Sports Park-based Waiheke United AFC adopted the Whole of Football Plan, which provides a comprehensive model with coherent pathways and age-appropriate skills development.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Waiheke United AFC_sentence_0",
"Waiheke United AFC_sentence_4",
"Waiheke United AFC_sentence_6"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2544,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.501",
"value": "?search="
},
{
"operation": "search",
"time": "21.127",
"value": "netangi Sports Park-based Waiheke United AFC"
},
{
"operation": "Now on",
"time": "22.501",
"value": "undefined"
},
{
"operation": "search",
"time": "37.319",
"value": "Onetangi Sports Park-based Waiheke United AFC"
},
{
"operation": "Now on",
"time": "37.997",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "42.473",
"value": "Waiheke United AFC"
},
{
"operation": "Now on",
"time": "43.118",
"value": "Waiheke United AFC"
},
{
"operation": "Highlighting",
"time": "46.335",
"value": "Waiheke United AFC_sentence_0"
},
{
"operation": "Page search",
"time": "69.43",
"value": "Whole of Football Plan"
},
{
"operation": "Highlighting",
"time": "77.406",
"value": "Waiheke United AFC_sentence_4"
},
{
"operation": "Highlighting",
"time": "85.351",
"value": "Waiheke United AFC_sentence_6"
},
{
"operation": "finish",
"time": "103.542",
"value": "finish"
}
]
}
|
exfever_8591
|
Black Sails is an American historical adventure television series created by a television producer, screenwriter, and director who co-created the tv series Jericho with an American screenwriter, film producer, film director, novelist, television writer, and television producer.
|
{
"evidence": [
"Black Sails (TV series)",
"Jonathan E. Steinberg",
"Stephen Chbosky"
],
"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:\nBlack Sails is an American historical adventure television series created by a television producer, screenwriter, and director who co-created the tv series Jericho with an American screenwriter, film producer, film director, novelist, television writer, and television producer.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Black Sails (TV series)",
"Jonathan E. Steinberg",
"Stephen Chbosky"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2545,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8139
|
Don Simpson was an American film producer, screenwriter, and actor who co-produced hit films, including a 1984 American buddy cop action comedy film directed by Eddie Murphy.
|
{
"evidence": [
"Don Simpson",
"Beverly Hills Cop"
],
"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:\nDon Simpson was an American film producer, screenwriter, and actor who co-produced hit films, including a 1984 American buddy cop action comedy film directed by Eddie Murphy.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Don Simpson",
"Beverly Hills Cop"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2546,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1009
|
Aaron Charles Carter's best-selling album was Aaron's Party, which was released in the year 2000.
|
{
"evidence": [
"Aaron Carter",
"Nickelodeon",
"QUBE"
],
"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:\nAaron Charles Carter's best-selling album was Aaron's Party, which was released in the year 2000.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Aaron Carter",
"Nickelodeon",
"QUBE"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2547,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_3201
|
Matthew Robert Smith is an English actor best known for his roles as the eleventh incarnation of the Doctor in a British science-fiction television program that depicts a life of a Doctor that explores the universe in a time-traveling spaceship called the time machine and spacecraft that is a registered trademark of the British Broadcasting Corporation.
|
{
"evidence": [
"Matt Smith (actor)",
"Doctor Who",
"TARDIS"
],
"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:\nMatthew Robert Smith is an English actor best known for his roles as the eleventh incarnation of the Doctor in a British science-fiction television program that depicts a life of a Doctor that explores the universe in a time-traveling spaceship called the time machine and spacecraft that is a registered trademark of the British Broadcasting Corporation.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Matt Smith (actor)",
"Doctor Who",
"TARDIS"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2548,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
56484
|
The United Nations Security Council Resolution 431 on Nambia's situation was adopted on 27 July 1978.
|
{
"evidence": [
"United Nations Security Council Resolution 431_sentence_0"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe United Nations Security Council Resolution 431 on Nambia's situation was adopted on 27 July 1978.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"United Nations Security Council Resolution 431_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2549,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.427",
"value": "?search="
},
{
"operation": "search",
"time": "14.546",
"value": "United Nations Security Council Resolution 431"
},
{
"operation": "Now on",
"time": "15.934",
"value": "United Nations Security Council Resolution 431"
},
{
"operation": "Highlighting",
"time": "26.989",
"value": "United Nations Security Council Resolution 431_sentence_0"
},
{
"operation": "finish",
"time": "45.569",
"value": "finish"
}
]
}
|
24486
|
India is Vanuatu's 5th largest export destination, with US$4.5 million worth of goods sent in the 2013-2014 year; which makes up 1.5% of all of Vanuatu's exports and is an increase from US$5.37 million in the 2012-2013 year.
|
{
"evidence": [
"India–Vanuatu relations_sentence_14"
],
"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:\nIndia is Vanuatu's 5th largest export destination, with US$4.5 million worth of goods sent in the 2013-2014 year; which makes up 1.5% of all of Vanuatu's exports and is an increase from US$5.37 million in the 2012-2013 year.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"India–Vanuatu relations_sentence_14"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2550,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.79",
"value": "?search="
},
{
"operation": "search",
"time": "48.849",
"value": "India–Vanuatu relations"
},
{
"operation": "Now on",
"time": "49.5",
"value": "India–Vanuatu relations"
},
{
"operation": "Highlighting",
"time": "54.544",
"value": "India–Vanuatu relations_sentence_14"
},
{
"operation": "finish",
"time": "170.61",
"value": "finish"
}
]
}
|
9069
|
Thomas Brinkwell was collated to a prebend in Lincoln Cathedral and several years later was nominated professor of divinity.
|
{
"evidence": [
"Thomas Brinknell_sentence_8",
"Thomas Brinknell_sentence_9"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThomas Brinkwell was collated to a prebend in Lincoln Cathedral and several years later was nominated professor of divinity.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Thomas Brinknell_sentence_8",
"Thomas Brinknell_sentence_9"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2551,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.516",
"value": "?search="
},
{
"operation": "search",
"time": "34.783",
"value": "Thomas Brinkwell"
},
{
"operation": "Now on",
"time": "35.394",
"value": "undefined"
},
{
"operation": "search",
"time": "42.723",
"value": "Thomas Brinkwell"
},
{
"operation": "Now on",
"time": "43.307",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "55.714",
"value": "thomas brinknell"
},
{
"operation": "Now on",
"time": "56.045",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "66.094",
"value": "Thomas Brinknell"
},
{
"operation": "Now on",
"time": "66.694",
"value": "Thomas Brinknell"
},
{
"operation": "Highlighting",
"time": "85.788",
"value": "Thomas Brinknell_sentence_8"
},
{
"operation": "Highlighting",
"time": "97.638",
"value": "Thomas Brinknell_sentence_9"
},
{
"operation": "finish",
"time": "137.358",
"value": "finish"
}
]
}
|
exfever_1055
|
O. J. Simpson was the fastest football player in Buffalo Bills.
|
{
"evidence": [
"O. J. Simpson",
"Running back"
],
"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:\nO. J. Simpson was the fastest football player in Buffalo Bills.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"O. J. Simpson",
"Running back"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2552,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1439
|
Dr. Dre ended his career as a member of an American electro group.
|
{
"evidence": [
"Dr. Dre",
"World Class Wreckin' Cru"
],
"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:\nDr. Dre ended his career as a member of an American electro group.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Dr. Dre",
"World Class Wreckin' Cru"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2553,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
22340
|
The model ALCO RS-3, a 1,600 hp (1.2 MW), B-B road switcher diesel-electric locomotive, fleet number 92-99 were acquired by Norfolk and Western Railways in 1956.
|
{
"evidence": [
"ALCO RS-3_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 model ALCO RS-3, a 1,600 hp (1.2 MW), B-B road switcher diesel-electric locomotive, fleet number 92-99 were acquired by Norfolk and Western Railways in 1956.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"ALCO RS-3_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2554,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.296",
"value": "?search="
},
{
"operation": "search",
"time": "52.747",
"value": "model ALCO RS-3"
},
{
"operation": "Now on",
"time": "54.98",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "60.255",
"value": "ALCO RS-3"
},
{
"operation": "Now on",
"time": "61.361",
"value": "ALCO RS-3"
},
{
"operation": "Highlighting",
"time": "86.964",
"value": "ALCO RS-3_sentence_0"
},
{
"operation": "Page search",
"time": "95.283",
"value": "92-99"
},
{
"operation": "Page search",
"time": "117.681",
"value": "fleet"
},
{
"operation": "Page search",
"time": "119.081",
"value": "fleet"
},
{
"operation": "Page search",
"time": "119.24",
"value": "fleet"
},
{
"operation": "Page search",
"time": "145.699",
"value": "Norfolk"
},
{
"operation": "Page search",
"time": "146.994",
"value": "Norfolk"
},
{
"operation": "Page search",
"time": "147.137",
"value": "Norfolk"
},
{
"operation": "Page search",
"time": "169.565",
"value": "western"
},
{
"operation": "Page search",
"time": "169.931",
"value": "western"
},
{
"operation": "Page search",
"time": "170.155",
"value": "western"
},
{
"operation": "Page search",
"time": "170.323",
"value": "western"
},
{
"operation": "finish",
"time": "191.595",
"value": "finish"
}
]
}
|
12676
|
On the list of Moody Yachts (owned by HanseYachts AG, Greifswald - the third largest production builder of sailing yachts globally) models, the fourth model, Moody 54 DS, is the longest and the heaviest.
|
{
"evidence": [
"HanseYachts_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:\nOn the list of Moody Yachts (owned by HanseYachts AG, Greifswald - the third largest production builder of sailing yachts globally) models, the fourth model, Moody 54 DS, is the longest and the heaviest.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"HanseYachts_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2555,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.898",
"value": "?search="
},
{
"operation": "search",
"time": "7.533",
"value": "Contains...list of Moody Yachts"
},
{
"operation": "Now on",
"time": "8.9",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "10.55",
"value": "Moody Yachts"
},
{
"operation": "Now on",
"time": "11.274",
"value": "Moody Yachts"
},
{
"operation": "Page search",
"time": "17.728",
"value": "HanseYachts"
},
{
"operation": "Highlighting",
"time": "28.111",
"value": "Moody Yachts_sentence_2"
},
{
"operation": "Highlighting",
"time": "28.112",
"value": "Moody Yachts_title"
},
{
"operation": "hyperlink",
"time": "35.439",
"value": "HanseYachts AG"
},
{
"operation": "Now on",
"time": "36.142",
"value": "HanseYachts"
},
{
"operation": "Page search",
"time": "39.235",
"value": "largest"
},
{
"operation": "Highlighting",
"time": "41.594",
"value": "HanseYachts_sentence_4"
},
{
"operation": "finish",
"time": "66.554",
"value": "finish"
}
]
}
|
67330
|
Pia Star, bred by and raced by Ada L. Rice and trained by Clyde Troutt, placed second in the 1965 Brooklyn and Suburban Handicaps and in 1966, the Widener Handicap at Hialeah Park Race Track in Florida.
|
{
"evidence": [
"Pia Star_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:\nPia Star, bred by and raced by Ada L. Rice and trained by Clyde Troutt, placed second in the 1965 Brooklyn and Suburban Handicaps and in 1966, the Widener Handicap at Hialeah Park Race Track in Florida.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Pia Star_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2556,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.063",
"value": "?search="
},
{
"operation": "search",
"time": "8.543",
"value": "Pia Star"
},
{
"operation": "Now on",
"time": "9.709",
"value": "Pia Star"
},
{
"operation": "Highlighting",
"time": "14.459",
"value": "Pia Star_sentence_2"
},
{
"operation": "Highlighting",
"time": "23.74",
"value": "Pia Star_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "25.032",
"value": "Pia Star_sentence_2"
},
{
"operation": "finish",
"time": "31.366",
"value": "finish"
}
]
}
|
75338
|
Marc Cheverie played for the Phoenix Coyotes after a trade in exhange for Justin Bernhardt on September of 2011 then he stayed with them for 3 more years.
|
{
"evidence": [
"Marc Cheverie_sentence_2",
"Marc Cheverie_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:\nMarc Cheverie played for the Phoenix Coyotes after a trade in exhange for Justin Bernhardt on September of 2011 then he stayed with them for 3 more years.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Marc Cheverie_sentence_2",
"Marc Cheverie_sentence_3"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2557,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.907",
"value": "?search="
},
{
"operation": "search",
"time": "9.3",
"value": "Marc Cheverie"
},
{
"operation": "Now on",
"time": "10.108",
"value": "Marc Cheverie"
},
{
"operation": "Highlighting",
"time": "48.198",
"value": "Marc Cheverie_sentence_2"
},
{
"operation": "Highlighting",
"time": "94.038",
"value": "Marc Cheverie_sentence_3"
},
{
"operation": "finish",
"time": "106.181",
"value": "finish"
}
]
}
|
exfever_7714
|
The Statue of Liberty is a colossal neoclassical sculpture on a federally owned island in Upper New York Bay in the United States.
|
{
"evidence": [
"Statue of Liberty",
"Liberty Island"
],
"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 Statue of Liberty is a colossal neoclassical sculpture on a federally owned island in Upper New York Bay in the United States.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Statue of Liberty",
"Liberty Island"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2558,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4494
|
Alia Bhatt is the son of an Indian film director, producer, and screenwriter known for his works in Hindi cinema.
|
{
"evidence": [
"Alia Bhatt",
"Mahesh Bhatt"
],
"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:\nAlia Bhatt is the son of an Indian film director, producer, and screenwriter known for his works in Hindi cinema.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Alia Bhatt",
"Mahesh Bhatt"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2559,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7792
|
Artemis is the goddess of the Earth's only natural satellite, which is smaller than any known small planetary-mass object.
|
{
"evidence": [
"Artemis",
"Moon",
"Dwarf planet"
],
"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:\nArtemis is the goddess of the Earth's only natural satellite, which is smaller than any known small planetary-mass object.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Artemis",
"Moon",
"Dwarf planet"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2560,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4669
|
Legion is an American superhero television series based on a character in a brand name and primary imprint of Marvel Worldwide Inc.
|
{
"evidence": [
"Legion (TV series)",
"Marvel Comics"
],
"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:\nLegion is an American superhero television series based on a character in a brand name and primary imprint of Marvel Worldwide Inc.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Legion (TV series)",
"Marvel Comics"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2561,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
42811
|
Yunyang District is a 3,863 km² district of the prefecture-level city of Shiyan in northwestern Guangzhou Province, China that was formerly known as Yun County or Yunxian,
|
{
"evidence": [
"Yunyang District_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:\nYunyang District is a 3,863 km² district of the prefecture-level city of Shiyan in northwestern Guangzhou Province, China that was formerly known as Yun County or Yunxian,\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Yunyang District_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2562,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.334",
"value": "?search="
},
{
"operation": "Now on",
"time": "1.77",
"value": "?search="
},
{
"operation": "search",
"time": "6.887",
"value": "Yunyang District"
},
{
"operation": "Now on",
"time": "7.979",
"value": "Yunyang District"
},
{
"operation": "Highlighting",
"time": "13.151",
"value": "Yunyang District_cell_0_5_1"
},
{
"operation": "Highlighting",
"time": "13.157",
"value": "Yunyang District_header_cell_0_5_0"
},
{
"operation": "Highlighting",
"time": "13.157",
"value": "Yunyang District_header_cell_0_4_0"
},
{
"operation": "Highlighting",
"time": "26.661",
"value": "Yunyang District_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "32.942",
"value": "Yunyang District_cell_0_5_1"
},
{
"operation": "Highlighting deleted",
"time": "32.948",
"value": "Yunyang District_header_cell_0_5_0"
},
{
"operation": "Highlighting deleted",
"time": "32.949",
"value": "Yunyang District_header_cell_0_4_0"
},
{
"operation": "finish",
"time": "42.922",
"value": "finish"
}
]
}
|
70445
|
Pyotr Ivanovich Sholokhov was born in 1798 and studied fine arts in Borisoglebsk.
|
{
"evidence": [
"Pyotr Ivanovich Sholokhov_sentence_5"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nPyotr Ivanovich Sholokhov was born in 1798 and studied fine arts in Borisoglebsk.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Pyotr Ivanovich Sholokhov_sentence_5"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2563,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.967",
"value": "?search="
},
{
"operation": "search",
"time": "21.728",
"value": "Pyotr Ivanovich Sholokhov"
},
{
"operation": "Now on",
"time": "22.446",
"value": "Pyotr Ivanovich Sholokhov"
},
{
"operation": "Highlighting",
"time": "29.336",
"value": "Pyotr Ivanovich Sholokhov_sentence_5"
},
{
"operation": "finish",
"time": "56.471",
"value": "finish"
}
]
}
|
exfever_4273
|
British Columbia is a western province that borders one of the three prairie provinces that are part of the Canadian region that includes the four western provinces.
|
{
"evidence": [
"British Columbia",
"Alberta",
"Western Canada"
],
"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:\nBritish Columbia is a western province that borders one of the three prairie provinces that are part of the Canadian region that includes the four western provinces.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"British Columbia",
"Alberta",
"Western Canada"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2564,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
46765
|
Plana Baixa is a comarca which is no longer an intermediate level of administrative subdivision between municipalities and provinces.
|
{
"evidence": [
"Plana Baixa_sentence_0",
"Comarques of the Valencian Community_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:\nPlana Baixa is a comarca which is no longer an intermediate level of administrative subdivision between municipalities and provinces.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Plana Baixa_sentence_0",
"Comarques of the Valencian Community_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2565,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.252",
"value": "?search="
},
{
"operation": "search",
"time": "48.615",
"value": "Plana Baixa"
},
{
"operation": "Now on",
"time": "49.996",
"value": "Plana Baixa"
},
{
"operation": "Highlighting",
"time": "78.938",
"value": "Plana Baixa_sentence_0"
},
{
"operation": "hyperlink",
"time": "105.359",
"value": "comarca"
},
{
"operation": "Now on",
"time": "106.632",
"value": "Comarques of the Valencian Community"
},
{
"operation": "Highlighting",
"time": "111.496",
"value": "Comarques of the Valencian Community_sentence_0"
},
{
"operation": "finish",
"time": "149.316",
"value": "finish"
}
]
}
|
exfever_906
|
That '70s Show is an American television period sitcom that aired on Fox from August 23, 1998, to May 18, 2006 that has teenage cast members including an American actor, model, producer, and entrepreneur best known for the film Coming Soon.
|
{
"evidence": [
"That '70s Show",
"Ashton Kutcher"
],
"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:\nThat '70s Show is an American television period sitcom that aired on Fox from August 23, 1998, to May 18, 2006 that has teenage cast members including an American actor, model, producer, and entrepreneur best known for the film Coming Soon.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"That '70s Show",
"Ashton Kutcher"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2566,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
54969
|
Pat Daly was an Irish footballer who joined the Shamrock Rovers in 1948 and only played with them.
|
{
"evidence": [
"Pat Daly_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:\nPat Daly was an Irish footballer who joined the Shamrock Rovers in 1948 and only played with them.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Pat Daly_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2567,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.938",
"value": "?search="
},
{
"operation": "Now on",
"time": "10.755",
"value": "?search="
},
{
"operation": "search",
"time": "51.566",
"value": "Pat Daly"
},
{
"operation": "Now on",
"time": "52.484",
"value": "Pat Daly"
},
{
"operation": "Highlighting",
"time": "87.677",
"value": "Pat Daly_sentence_0"
},
{
"operation": "Highlighting",
"time": "128.922",
"value": "Pat Daly_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "146.69",
"value": "Pat Daly_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "148.018",
"value": "Pat Daly_sentence_0"
},
{
"operation": "Highlighting",
"time": "149.042",
"value": "Pat Daly_sentence_2"
},
{
"operation": "finish",
"time": "185.058",
"value": "finish"
}
]
}
|
52000
|
Wesley Barresi was named in the Netherlands' One Month International (ODI) squad; he was also named as the substitute player for Poland.
|
{
"evidence": [
"Wesley Barresi_sentence_3",
"Wesley Barresi_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:\nWesley Barresi was named in the Netherlands' One Month International (ODI) squad; he was also named as the substitute player for Poland.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Wesley Barresi_sentence_3",
"Wesley Barresi_sentence_4"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2568,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.732",
"value": "?search="
},
{
"operation": "search",
"time": "9.575",
"value": "Wesley Barresi"
},
{
"operation": "Now on",
"time": "10.542",
"value": "Wesley Barresi"
},
{
"operation": "Highlighting",
"time": "18.023",
"value": "Wesley Barresi_sentence_3"
},
{
"operation": "hyperlink",
"time": "78.424",
"value": "inaugural edition"
},
{
"operation": "Now on",
"time": "79.395",
"value": "2019 Euro T20 Slam"
},
{
"operation": "back-button-clicked",
"time": "102.894",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "103.044",
"value": "Wesley Barresi"
},
{
"operation": "Highlighting",
"time": "109.755",
"value": "Wesley Barresi_sentence_4"
},
{
"operation": "Highlighting",
"time": "114.513",
"value": "Wesley Barresi_sentence_5"
},
{
"operation": "Highlighting deleted",
"time": "116.99",
"value": "Wesley Barresi_sentence_5"
},
{
"operation": "finish",
"time": "124.256",
"value": "finish"
}
]
}
|
2467
|
Billy Colewas, a former British track and field athlete was a member of the Newham and Essex Beagles athletics club where he won the 1980 under-17s AAA indoor title.
|
{
"evidence": [
"Billy Cole_sentence_1",
"Billy Cole_sentence_7",
"Billy Cole_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:\nBilly Colewas, a former British track and field athlete was a member of the Newham and Essex Beagles athletics club where he won the 1980 under-17s AAA indoor title.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Billy Cole_sentence_1",
"Billy Cole_sentence_7",
"Billy Cole_sentence_8"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2569,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "search",
"time": "34.357",
"value": "Billy Colewas"
},
{
"operation": "search",
"time": "64.145",
"value": "Billy Cole"
},
{
"operation": "Highlighting",
"time": "68.895",
"value": "Billy Cole_sentence_1"
},
{
"operation": "Highlighting",
"time": "78.622",
"value": "Billy Cole_sentence_7"
},
{
"operation": "Highlighting",
"time": "83.558",
"value": "Billy Cole_sentence_8"
},
{
"operation": "finish",
"time": "94.843",
"value": "finish"
}
]
}
|
exfever_4045
|
Metallica was one of the founding big four bands, like the ones from California and New York.
|
{
"evidence": [
"Metallica",
"Slayer",
"Anthrax (American band)"
],
"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:\nMetallica was one of the founding big four bands, like the ones from California and New York.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Metallica",
"Slayer",
"Anthrax (American band)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2570,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_286
|
The Olympic Games are normally held every four years, alternating between the Summer Olympics and the international?sports event held on snow or ice, and feature organized sporting events, often held over multiple days, featuring competition in many different sports among organized teams of athletes from nation-states.
|
{
"evidence": [
"Olympic Games",
"Winter Olympic Games",
"Multi-sport event"
],
"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 Olympic Games are normally held every four years, alternating between the Summer Olympics and the international?sports event held on snow or ice, and feature organized sporting events, often held over multiple days, featuring competition in many different sports among organized teams of athletes from nation-states.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Olympic Games",
"Winter Olympic Games",
"Multi-sport event"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2571,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
31962
|
Sampangan, where a charity event known as the "terminal" is celebrated, is a village in the Gajahmungkur district of Semarang City.
|
{
"evidence": [
"Sampangan_sentence_3",
"Sampangan_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:\nSampangan, where a charity event known as the \"terminal\" is celebrated, is a village in the Gajahmungkur district of Semarang City.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Sampangan_sentence_3",
"Sampangan_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2572,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.782",
"value": "?search="
},
{
"operation": "search",
"time": "48.789",
"value": "Sampangan"
},
{
"operation": "Now on",
"time": "49.66",
"value": "Sampangan"
},
{
"operation": "Highlighting",
"time": "68.334",
"value": "Sampangan_sentence_3"
},
{
"operation": "Highlighting",
"time": "80.69",
"value": "Sampangan_sentence_0"
},
{
"operation": "finish",
"time": "144.459",
"value": "finish"
}
]
}
|
28664
|
The 2018 Australian federal budget, presented by then Treasurer Scott Morrison who later became Prime Minister, allowed for the Social Welfare Debt Recovery scheme to be extended for another three years but has noted no increase in the Newstart Allowance.
|
{
"evidence": [
"2018 Australian federal budget_sentence_1",
"Scott Morrison_sentence_1",
"2018 Australian federal budget_sentence_19",
"2018 Australian federal budget_sentence_22"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe 2018 Australian federal budget, presented by then Treasurer Scott Morrison who later became Prime Minister, allowed for the Social Welfare Debt Recovery scheme to be extended for another three years but has noted no increase in the Newstart Allowance.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"2018 Australian federal budget_sentence_1",
"Scott Morrison_sentence_1",
"2018 Australian federal budget_sentence_19",
"2018 Australian federal budget_sentence_22"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2573,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.088",
"value": "?search="
},
{
"operation": "search",
"time": "6.85",
"value": "2018 Australian federal budget"
},
{
"operation": "Now on",
"time": "7.695",
"value": "2018 Australian federal budget"
},
{
"operation": "Highlighting",
"time": "11.314",
"value": "2018 Australian federal budget_sentence_1"
},
{
"operation": "hyperlink",
"time": "16.287",
"value": "Scott Morrison"
},
{
"operation": "Now on",
"time": "16.998",
"value": "Scott Morrison"
},
{
"operation": "Highlighting",
"time": "19.595",
"value": "Scott Morrison_sentence_1"
},
{
"operation": "back-button-clicked",
"time": "23.713",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "23.842",
"value": "2018 Australian federal budget"
},
{
"operation": "Highlighting",
"time": "37.147",
"value": "2018 Australian federal budget_sentence_19"
},
{
"operation": "Highlighting",
"time": "47.845",
"value": "2018 Australian federal budget_sentence_22"
},
{
"operation": "finish",
"time": "58.236",
"value": "finish"
}
]
}
|
exfever_2808
|
Abby Sciuto was introduced in two episodes of the television show created by Donald P. Bellisario.
|
{
"evidence": [
"Abby Sciuto",
"JAG (TV series)"
],
"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:\nAbby Sciuto was introduced in two episodes of the television show created by Donald P. Bellisario.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Abby Sciuto",
"JAG (TV series)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2574,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1909
|
Wrath of the Titans is a 2013 action fantasy film that stars a British actress who began her acting career by appearing in stage productions such as Romeo and Juliet.
|
{
"evidence": [
"Wrath of the Titans",
"Rosamund Pike"
],
"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:\nWrath of the Titans is a 2013 action fantasy film that stars a British actress who began her acting career by appearing in stage productions such as Romeo and Juliet.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Wrath of the Titans",
"Rosamund Pike"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2575,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5582
|
Gorillaz are the most prominent British?cartoon group.
|
{
"evidence": [
"Demon Days",
"Virtual band"
],
"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:\nGorillaz are the most prominent British?cartoon group.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Demon Days",
"Virtual band"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2576,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7262
|
The Twilight Saga is a series of three fantasy romance novels written by Stephanie Meyer, with the theme that literature covers the spectrum of literary works concerned with bloodsuckers.
|
{
"evidence": [
"Twilight (novel series)",
"Vampire literature"
],
"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 Twilight Saga is a series of three fantasy romance novels written by Stephanie Meyer, with the theme that literature covers the spectrum of literary works concerned with bloodsuckers.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Twilight (novel series)",
"Vampire literature"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2577,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6184
|
Sansa Stark is a fictional character in an American fantasy drama television series created by D. B. Weiss and an American writer, director, and producer, known for writing 25th Hour.
|
{
"evidence": [
"Sansa Stark",
"Game of Thrones",
"David Benioff"
],
"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:\nSansa Stark is a fictional character in an American fantasy drama television series created by D. B. Weiss and an American writer, director, and producer, known for writing 25th Hour.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Sansa Stark",
"Game of Thrones",
"David Benioff"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2578,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
29746
|
Bernard Pagel obtained his Ph.D. in Cambridge in 1955 and became a Research Fellow at Sidney Sussex college in 1953-1956.
|
{
"evidence": [
"Bernard Pagel_sentence_3",
"Bernard Pagel_sentence_4"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nBernard Pagel obtained his Ph.D. in Cambridge in 1955 and became a Research Fellow at Sidney Sussex college in 1953-1956.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Bernard Pagel_sentence_3",
"Bernard Pagel_sentence_4"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2579,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.642",
"value": "?search="
},
{
"operation": "search",
"time": "3.896",
"value": "Bernard Pagel"
},
{
"operation": "Now on",
"time": "4.576",
"value": "Bernard Pagel"
},
{
"operation": "Highlighting",
"time": "18.904",
"value": "Bernard Pagel_sentence_3"
},
{
"operation": "Highlighting",
"time": "22.392",
"value": "Bernard Pagel_sentence_4"
},
{
"operation": "finish",
"time": "30.227",
"value": "finish"
}
]
}
|
exfever_4993
|
Little Miss Sunshine is a 2006 American tragicomedy road film and the feature film directorial debut of the husband–wife team of American film and the head of music video production.
|
{
"evidence": [
"Little Miss Sunshine",
"Jonathan Dayton and Valerie Faris",
"Music video director"
],
"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:\nLittle Miss Sunshine is a 2006 American tragicomedy road film and the feature film directorial debut of the husband–wife team of American film and the head of music video production.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Little Miss Sunshine",
"Jonathan Dayton and Valerie Faris",
"Music video director"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2580,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1254
|
The Damages was the first American legal thriller television series that premiered on FX.
|
{
"evidence": [
"Damages (TV series)",
"Ellen Parsons"
],
"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 Damages was the first American legal thriller television series that premiered on FX.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Damages (TV series)",
"Ellen Parsons"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2581,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
31589
|
The B61’s route is between Long Island City (a residential and commercial neighborhood on the west side of Queens, New York City) and the south end of Van Brunt Street in Red Hook is still in existence and has never been changed.
|
{
"evidence": [
"B61 and B62 buses_sentence_23"
],
"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 B61’s route is between Long Island City (a residential and commercial neighborhood on the west side of Queens, New York City) and the south end of Van Brunt Street in Red Hook is still in existence and has never been changed.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"B61 and B62 buses_sentence_23"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 2582,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.62",
"value": "?search="
},
{
"operation": "search",
"time": "16.637",
"value": "B61"
},
{
"operation": "Now on",
"time": "17.497",
"value": "B61"
},
{
"operation": "hyperlink",
"time": "24.342",
"value": "B61 (New York City bus)"
},
{
"operation": "Now on",
"time": "24.855",
"value": "B61 and B62 buses"
},
{
"operation": "back-button-clicked",
"time": "73.293",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "73.382",
"value": "B61"
},
{
"operation": "hyperlink",
"time": "81.967",
"value": "B61 (New York City bus)"
},
{
"operation": "Now on",
"time": "82.38",
"value": "B61 (New York City bus)"
},
{
"operation": "Highlighting",
"time": "150.873",
"value": "B61 and B62 buses_sentence_3"
},
{
"operation": "hyperlink",
"time": "154.453",
"value": "Long Island City"
},
{
"operation": "Now on",
"time": "155.39",
"value": "Long Island City, Queens"
},
{
"operation": "Highlighting",
"time": "159.234",
"value": "Long Island City_sentence_0"
},
{
"operation": "Highlighting",
"time": "159.237",
"value": "Long Island City_title"
},
{
"operation": "back-button-clicked",
"time": "164.473",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "164.984",
"value": "B61 and B62 buses"
},
{
"operation": "Highlighting",
"time": "256.074",
"value": "B61 and B62 buses_sentence_23"
},
{
"operation": "hyperlink",
"time": "258.665",
"value": "New York City"
},
{
"operation": "Now on",
"time": "259.762",
"value": "New York City"
},
{
"operation": "back-button-clicked",
"time": "260.563",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "260.676",
"value": "B61 and B62 buses"
},
{
"operation": "finish",
"time": "285.177",
"value": "finish"
}
]
}
|
71333
|
The governmental learning spiral is a technique facilitated by a learning broker which include logistics, performance , drafting and implementation of the modules, moderation of the learning sessions.
|
{
"evidence": [
"Governmental learning spiral_sentence_0",
"Governmental learning spiral_sentence_4",
"Governmental learning spiral_sentence_5"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe governmental learning spiral is a technique facilitated by a learning broker which include logistics, performance , drafting and implementation of the modules, moderation of the learning sessions.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Governmental learning spiral_sentence_0",
"Governmental learning spiral_sentence_4",
"Governmental learning spiral_sentence_5"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 2583,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.746",
"value": "?search="
},
{
"operation": "search",
"time": "16.55",
"value": "Governmental learning spiral"
},
{
"operation": "Now on",
"time": "17.248",
"value": "Governmental learning spiral"
},
{
"operation": "Highlighting",
"time": "34.259",
"value": "Governmental learning spiral_sentence_0"
},
{
"operation": "Highlighting",
"time": "47.675",
"value": "Governmental learning spiral_sentence_4"
},
{
"operation": "Highlighting",
"time": "79.27",
"value": "Governmental learning spiral_sentence_5"
},
{
"operation": "Page search",
"time": "132.25",
"value": "performance"
},
{
"operation": "finish",
"time": "221.269",
"value": "finish"
}
]
}
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.