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 |
|---|---|---|---|---|---|---|---|---|
73222
|
Brewer, born in 1810, in Norwich to a Baptist schoolmaster, was matriculated at Queen's College, Oxford in 1827, before being ordained in the Church of England in 1837.
|
{
"evidence": [
"John Sherren Brewer_sentence_0",
"John Sherren Brewer_sentence_3",
"John Sherren Brewer_sentence_4",
"John Sherren Brewer_sentence_5"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information above.\n\n<think>\n...\n</think>\n- Use for every piece of reasoning.\n- During reasoning, you must verify the claim step by step based on the given information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer>\n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\nVerify the claim:\nBrewer, born in 1810, in Norwich to a Baptist schoolmaster, was matriculated at Queen's College, Oxford in 1827, before being ordained in the Church of England in 1837.\n<information>\n[[John Sherren Brewer_sentence_0]]: John Sherren Brewer (1810 – February 1879) was an English clergyman, historian and scholar.\n[[John Sherren Brewer_sentence_1]]: He was a brother of [[E._Cobham_Brewer|E.\n[[John Sherren Brewer_sentence_3]]: Brewer was born in [[Norwich,_England|Norwich]], the son of a [[Baptist|Baptist]] schoolmaster.\n[[John Sherren Brewer_sentence_4]]: He matriculated at [[The_Queen's_College,_Oxford|Queen's College, Oxford]] in 1827, graduating B.A. in 1833, M.A. 1835.\n[[John Sherren Brewer_sentence_5]]: He was ordained in the Church of England in 1837, and became chaplain to a central London [[Workhouse|workhouse]].\n[[John Sherren Brewer_sentence_6]]: In 1839 he was appointed lecturer in classical literature at [[King's_College_London|King's College London]], and in 1858 he became professor of English language and [[English_literature|literature]] and lecturer in modern history, succeeding [[John_Frederick_Denison_Maurice|FD Maurice]].\n[[John Sherren Brewer_sentence_8]]: From 1854 onwards, Brewer combined these duties with [[Journalism|journalistic]] work on the [[Morning_Herald|Morning Herald]], [[Morning_Post|Morning Post]] and [[Evening_Standard|Standard]].\n[[John Sherren Brewer_sentence_10]]: He was also made Reader, and subsequently Preacher, at the [[Rolls_Chapel|Rolls Chapel]].\n[[John Sherren Brewer_sentence_13]]: There Brewer had time to continue his task of preparing his Letters and Papers of the Reign of Henry VIII, the \"Introductions\" to which (published separately, under the title The Reign of Henry VIII, in 1884) form a scholarly and authoritative history of [[Henry_VIII_of_England|Henry VIII]]'s reign.\n</information>\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"John Sherren Brewer_sentence_0",
"John Sherren Brewer_sentence_3",
"John Sherren Brewer_sentence_4",
"John Sherren Brewer_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 8176,
"split": "train_offline"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.889",
"value": "?search="
},
{
"operation": "search",
"time": "13.743",
"value": "Brewer"
},
{
"operation": "Now on",
"time": "16.448",
"value": "Brewing"
},
{
"operation": "hyperlink",
"time": "85.467",
"value": "Brewer (disambiguation)"
},
{
"operation": "Now on",
"time": "86.028",
"value": "Brewer (disambiguation)"
},
{
"operation": "hyperlink",
"time": "102.897",
"value": "Brewer (surname)"
},
{
"operation": "Now on",
"time": "103.42",
"value": "Brewer (surname)"
},
{
"operation": "Page search",
"time": "112.826",
"value": "1810"
},
{
"operation": "hyperlink",
"time": "130.81",
"value": "John Sherren Brewer"
},
{
"operation": "Now on",
"time": "131.376",
"value": "John Sherren Brewer"
},
{
"operation": "Highlighting",
"time": "212.248",
"value": "John Sherren Brewer_sentence_0"
},
{
"operation": "Highlighting",
"time": "216.405",
"value": "John Sherren Brewer_sentence_3"
},
{
"operation": "Highlighting",
"time": "222.117",
"value": "John Sherren Brewer_sentence_4"
},
{
"operation": "Highlighting",
"time": "226.373",
"value": "John Sherren Brewer_sentence_5"
},
{
"operation": "finish",
"time": "240.068",
"value": "finish"
}
]
}
|
exfever_4778
|
The Armenian genocide was the systematic destruction of the Armenian people in the empire that controlled much of Southeast Europe.
|
{
"evidence": [
"Armenian Genocide",
"Ottoman Empire"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information above.\n\n<think>\n...\n</think>\n- Use for every piece of reasoning.\n- During reasoning, you must verify the claim step by step based on the given information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer>\n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\nVerify the claim:\nThe Armenian genocide was the systematic destruction of the Armenian people in the empire that controlled much of Southeast Europe.\n<information>\n[[Armenian Genocide]]: The Armenian Genocide (Armenian: Հայոց ցեղասպանություն , \"Hayots tseghaspanutyun\"), also known as the Armenian Holocaust, was the Ottoman government's systematic extermination of 1.5 million Armenians, mostly Ottoman citizens within the Ottoman Empire and its successor state, the Republic of Turkey. The starting date is conventionally held to be 24 April 1915, the day that Ottoman authorities rounded up, arrested, and deported 235 to 270 Armenian intellectuals and community leaders from Constantinople to the region of Ankara, the majority of whom were eventually murdered. The genocide was carried out during and after World War I and implemented in two phases—the wholesale killing of the able-bodied male population through massacre and subjection of army conscripts to forced labour, followed by the deportation of women, children, the elderly, and the infirm on death marches leading to the Syrian Desert. Driven forward by military escorts, the deportees were deprived of food and water and subjected to periodic robbery, rape, and massacre. Other indigenous and Christian ethnic groups, such as the Assyrians and the Ottoman Greeks, were similarly targeted for extermination by the Ottoman government in the Assyrian genocide and the Greek genocide, and their treatment is considered by some historians to be part of the same genocidal policy. Most Armenian diaspora communities around the world came into being as a direct result of the genocide.\n\n[[Rover's Return (album)]]: Rover's Return is the fourth solo studio album from English musician John Waite, released in 1987.\n\n[[Polmacoxib]]: Polmacoxib (Acelex) is a nonsteroidal anti-inflammatory drug (NSAID) used to treat osteoarthritis. It was developed as CG100649 and approved for use in South Korea in February 2015. It inhibits the enzymes carbonic anhydrase and COX-2.\n\n[[Ottoman Empire]]: The Ottoman Empire (; Ottoman Turkish: دولت عليه عثمانیه , \"Devlet-i ʿAlīye-i ʿOsmānīye \", literally The Exalted Ottoman State; Modern Turkish: \"Osmanlı İmparatorluğu \" or \"Osmanlı Devleti \"; also historically known in Western Europe as the Turkish Empire, Ottoman Turkey or simply Turkey) was an empire founded at the end of the 13th century in northwestern Anatolia in the town of Söğüt (modern-day Bicelik Province) by the Oghuz Turkish tribal leader Osman. After 1354, the Ottomans crossed into Europe, and with the conquest of the Balkans, the Ottoman Beylik was transformed into a transcontinental empire. The Ottomans ended the Byzantine Empire with the 1453 conquest of Constantinople by Mehmed the Conqueror.\n\n[[Go (Motion City Soundtrack album)]]: Go is the fifth studio album by American rock band Motion City Soundtrack. Produced by Ed Ackerson and the band themselves, the album was released on June 12, 2012, in the United States by Epitaph Records and the band's own label, The Boombox Generation. Previously, the group had released their major-label debut, \"My Dinosaur Life\" (2010) on Columbia; the band parted ways with the label due to the record's lackluster commercial response. Following this, the band returned to their home of Minneapolis, Minnesota, with producer and longtime friend Ed Ackerson. The band entered a local studio and recorded their next effort on their own time and finances. The band recorded \"Go\" without a label, and put it upon themselves to pay for mixing and mastering and have conversations with distributors later.\n\n[[Vacations in Majorca]]: Vacations in Majorca (Italian: \"Brevi amori a Palma di Majorca\" ) is a 1959 Italian comedy film directed by Giorgio Bianchi.\n</information>\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Armenian Genocide",
"Ottoman Empire"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 8177,
"split": "train_offline"
}
|
{
"annotator_operations": []
}
|
16330
|
Major General Jim Muhwezi (born August 23, 1950 in Uganda) who is currently Uganda's Minister of Information & National Guidance, graduated from Makerere University with a degree of Bachelor of Laws, and Tanzania Police Academy where he finished the Police Cadet Course.
|
{
"evidence": [
"Jim Muhwezi_sentence_1"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information above.\n\n<think>\n...\n</think>\n- Use for every piece of reasoning.\n- During reasoning, you must verify the claim step by step based on the given information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer>\n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\nVerify the claim:\nMajor General Jim Muhwezi (born August 23, 1950 in Uganda) who is currently Uganda's Minister of Information & National Guidance, graduated from Makerere University with a degree of Bachelor of Laws, and Tanzania Police Academy where he finished the Police Cadet Course.\n<information>\n[[Jim Muhwezi_sentence_1]]: He is the former [[Information|Minister of Information and National Guidance]] in the [[Cabinet_of_Uganda|Ugandan Cabinet]].\n[[Jim Muhwezi_sentence_6]]: He also has received Senior Military Police and Security Intelligence Training from [[Tanzania|Tanzania]] and the former [[Soviet_Union|Soviet Union]].\n[[Jim Muhwezi_sentence_11]]: During the same period, he concurrently served as the first Director General of the [[Internal_Security_Organisation|Internal Security Organisation]] (ISO).\n[[Jim Muhwezi_sentence_13]]: During his ten-year tenure at the agency, no allegations of [[Torture|torture]] were raised against the agency.\n</information>\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Jim Muhwezi_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 8178,
"split": "train_offline"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.639",
"value": "?search="
},
{
"operation": "search",
"time": "20.973",
"value": "Major General Jim Muhwezi"
},
{
"operation": "Now on",
"time": "22.926",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "25.834",
"value": "Jim Muhwezi"
},
{
"operation": "Now on",
"time": "26.506",
"value": "Jim Muhwezi"
},
{
"operation": "Highlighting",
"time": "29.46",
"value": "Jim Muhwezi_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "40.779",
"value": "Jim Muhwezi_sentence_0"
},
{
"operation": "Highlighting",
"time": "55.023",
"value": "Jim Muhwezi_sentence_1"
},
{
"operation": "finish",
"time": "79.123",
"value": "finish"
}
]
}
|
43995
|
At the age of 18, a year before her dad died, Blaire White was a victim of date rape by a man, which she later revealed in a video collaboration with Shane Dawson.
|
{
"evidence": [
"Blaire White_sentence_5",
"Blaire White_sentence_6"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information above.\n\n<think>\n...\n</think>\n- Use for every piece of reasoning.\n- During reasoning, you must verify the claim step by step based on the given information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer>\n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\nVerify the claim:\nAt the age of 18, a year before her dad died, Blaire White was a victim of date rape by a man, which she later revealed in a video collaboration with Shane Dawson.\n<information>\n[[Blaire White_sentence_5]]: At the age of 18, she was a victim of date rape by a man, which she later revealed in a video collaboration with [[Shane_Dawson|Shane Dawson]].\n[[Blaire White_sentence_6]]: White's father died the following year.\n[[Blaire White_sentence_7]]: When she was 20 years old, she came out as transgender to her family and friends and began [[Feminizing_hormone_therapy|feminizing hormone therapy]] in 2015.\n[[Blaire White_sentence_8]]: White attended [[California_State_University,_Chico|California State University, Chico]], where she studied computer science.\n[[Blaire White_sentence_11]]: She continued making videos about her views on social issues like [[Gender_politics|gender politics]].\n[[Blaire White_sentence_24]]: In October 2020, White received criticism and backlash following a video she posted about transgender athlete [[Janae_Marie_Kroc|Janae Marie Kroc]].\n[[Blaire White_sentence_30]]: She describes her political beliefs as [[Centre-right_politics|center-right]] and is non-religious, stating that religion has had little significance in her life.\n[[Blaire White_sentence_32]]: In June 2018, White announced her engagement to fellow YouTuber Joey Sarson.\n</information>\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Blaire White_sentence_5",
"Blaire White_sentence_6"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 8179,
"split": "train_offline"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.712",
"value": "?search="
},
{
"operation": "search",
"time": "22.805",
"value": "Blaire White"
},
{
"operation": "Now on",
"time": "23.884",
"value": "Blaire White"
},
{
"operation": "Highlighting",
"time": "41.21",
"value": "Blaire White_sentence_5"
},
{
"operation": "Highlighting",
"time": "59.201",
"value": "Blaire White_sentence_6"
},
{
"operation": "finish",
"time": "440.06",
"value": "finish"
}
]
}
|
exfever_5794
|
Wolfgang Amadeus Mozart was the first prolific and influential composer of the Classical period.
|
{
"evidence": [
"Wolfgang Amadeus Mozart",
"Choir"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information above.\n\n<think>\n...\n</think>\n- Use for every piece of reasoning.\n- During reasoning, you must verify the claim step by step based on the given information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer>\n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\nVerify the claim:\nWolfgang Amadeus Mozart was the first prolific and influential composer of the Classical period.\n<information>\n[[List of state leaders in 1013]]: This is a list of heads of state, government leaders, and other rulers in the year .\n\n[[Choir]]: A choir ( ) (also known as a quire, chorale or chorus) is a musical ensemble of singers. Choral music, in turn, is the music written specifically for such an ensemble to perform. Choirs may perform music from the classical music repertoire, which spans from the Medieval era to the present, or popular music repertoire. Most choirs are led by a conductor, who leads the performances with arm and face gestures.\n\n[[1948 NCAA Cross Country Championships]]: The 1948 NCAA Cross Country Championships were the tenth annual cross country meet to determine the team and individual national champions of men's collegiate cross country running in the United States.\n\n[[List of United States Supreme Court cases, volume 349]]: List of United States Supreme Court cases, volume 349\n\n[[Wolfgang Amadeus Mozart]]: Wolfgang Amadeus Mozart ( ; ; ] ; 27 January 1756 – 5 December 1791), baptised as Johannes Chrysostomus Wolfgangus Theophilus Mozart, was a prolific and influential composer of the Classical era.\n\n[[Portal Bridge]]: The Portal Bridge is a railroad bridge over the Hackensack River in northeastern New Jersey, United States, just west of Secaucus Junction. It is a two-track, moveable swing-span between the towns of Kearny and Secaucus owned and operated by Amtrak as part of the Northeast Corridor and also used by New Jersey Transit. s of 2015 , it was considered the busiest train span in the Western Hemisphere, carrying between 150,000 and 200,000 passengers per day.\n</information>\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Wolfgang Amadeus Mozart",
"Choir"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 8180,
"split": "train_offline"
}
|
{
"annotator_operations": []
}
|
exfever_2043
|
Elizabeth Hughes Meriwether is known for creating an Asian television sitcom for the Fox Broadcasting Company that stars an American actor, comedian, film producer, and screenwriter.
|
{
"evidence": [
"Elizabeth Meriwether",
"New Girl",
"Jake Johnson"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information above.\n\n<think>\n...\n</think>\n- Use for every piece of reasoning.\n- During reasoning, you must verify the claim step by step based on the given information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer>\n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\nVerify the claim:\nElizabeth Hughes Meriwether is known for creating an Asian television sitcom for the Fox Broadcasting Company that stars an American actor, comedian, film producer, and screenwriter.\n<information>\n[[Jake Johnson]]: Jake Johnson (born Mark Jake Johnson Weinberger; May 28, 1978) is an American actor, comedian and director, most commonly known for playing Nick Miller in the Fox sitcom \"New Girl\" opposite Zooey Deschanel, for which he has received a Teen Choice Award nomination among others. Johnson also co-starred in the 2009 film \"Paper Heart\" and the 2012 film \"Safety Not Guaranteed\", as well as appearing in \"Get Him to the Greek\", \"21 Jump Street\", and \"Jurassic World\". His first starring role in a feature film was \"Drinking Buddies\", and he also starred in the 2014 comedy \"Let's Be Cops\", alongside fellow \"New Girl\" star Damon Wayans, Jr.\n\n[[Wrestling at the 2011 Pan American Games – Men's Greco-Roman 96 kg]]: Wrestling at the 2011 Pan American Games – Men's Greco-Roman 96 kg\n\n[[Hurones]]: Hurones is a municipality located in the province of Burgos, Castile and León, Spain. According to the 2004 census (INE), the municipality has a population of 70 inhabitants.\n\n[[New Girl]]: New Girl is an American sitcom television series that premiered on Fox on September 20, 2011. Developed by Elizabeth Meriwether under the working title \"Chicks & Dicks\", the series revolves around a kooky teacher, Jess (Zooey Deschanel), after she moves into a Los Angeles loft with three men, Nick (Jake Johnson), Schmidt (Max Greenfield), and Winston (Lamorne Morris); Jess' best friend Cece (Hannah Simone) and on-again-off-again loft mate Coach (Damon Wayans Jr.) also appear regularly. The show combines comedy and drama elements as the characters, who are in their early thirties, deal with maturing relationships and career choices. The sixth and most current season premiered on September 20, 2016.\n\n[[Elizabeth Meriwether]]: Elizabeth Hughes Meriwether (born 1981) is an American playwright and screenwriter, television showrunner, producer, and writer. She is known for the 2010 play \"Oliver Parker!\", the 2011 romantic comedy film \"No Strings Attached,\" and the sitcom \"New Girl\" for Fox, starring Zooey Deschanel.\n\n[[When Yankee Doodle Learns to "Parlez Vous Français"]]: When Yankee Doodle Learns to "Parlez Vous Français"\n</information>\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Elizabeth Meriwether",
"New Girl",
"Jake Johnson"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 8181,
"split": "train_offline"
}
|
{
"annotator_operations": []
}
|
73593
|
During the 1938-39 Lancashire Cup Final, Jim Sullivan of Wigan and Harold Osbaldestin of Salford both played the fullback position for their respective teams.
|
{
"evidence": [
"Jim Sullivan (Welsh rugby league)_sentence_17",
"Harold Osbaldestin_sentence_6"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information above.\n\n<think>\n...\n</think>\n- Use for every piece of reasoning.\n- During reasoning, you must verify the claim step by step based on the given information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer>\n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\nVerify the claim:\nDuring the 1938-39 Lancashire Cup Final, Jim Sullivan of Wigan and Harold Osbaldestin of Salford both played the fullback position for their respective teams.\n<information>\n[[Harold Osbaldestin_sentence_2]]: Harold Osbaldestin was born in [[Whelley|Whelley]], [[Wigan|Wigan]], [[Lancashire|Lancashire]], England, and he died aged 46 in [[Swinton,_Greater_Manchester|Swinton]], [[Lancashire|Lancashire]], England.\n[[Harold Osbaldestin_sentence_3]]: Osbaldestin was one of the players who successfully toured in France with Salford in 1934, during which the Salford team earned the name \"Les Diables Rouges\", the seventeen players were; [[Joe_Bradbury|Joe Bradbury]], [[Bob_Brown_(rugby_league)|Bob Brown]], [[Aubrey_Casewell|Aubrey Casewell]], [[Paddy_Dalton|Paddy Dalton]], [[Bert_Day|Bert Day]], [[Cliff_Evans_(rugby_league)|Cliff Evans]], [[Jack_Feetham|Jack Feetham]], George Harris, [[Barney_Hudson|Barney Hudson]], [[Emlyn_Jenkins|Emlyn Jenkins]], [[Alf_Middleton|Alf Middleton]], [[Sammy_Miller_(rugby_league)|Sammy Miller]], Osbaldestin, [[Les_Pearson|Les Pearson]], [[Gus_Risman|Gus Risman]], [[Billy_Watkins_(rugby)|Billy Watkins]] and [[Billy_Williams_(rugby,_born_1905)|Billy Williams]].\n[[Harold Osbaldestin_sentence_5]]: Osbaldestin played [[Fullback_(rugby_league)|fullback]] in Salford's 7–4 victory over [[Barrow_Raiders|Barrow]] in the [[1938_Challenge_Cup|1938]] [[Challenge_Cup|Challenge Cup]] Final during the [[1937–38_Northern_Rugby_Football_League_season#Challenge_Cup|1937–38 season]] at [[Wembley_Stadium_(1923)|Wembley Stadium]], London, on Saturday 7 May 1938.\n[[Harold Osbaldestin_sentence_6]]: About Osbaldestin's time, there was Salford's 10–8 victory over Swinton in the [[1931_Lancashire_Cup|1931]] [[RFL_Lancashire_Cup|Lancashire]] [[Rugby_league_county_cups|County Cup]] Final during the [[1931–32_Northern_Rugby_Football_League_season|1931–32 season]] at [[The_Cliff_(training_ground)|The Cliff]], [[Broughton,_Salford|Broughton, Salford]] on Saturday 21 November 1931, the 21–12 victory over [[Wigan_Warriors|Wigan]] in the [[1934_Lancashire_Cup|1934]] [[RFL_Lancashire_Cup|Lancashire]] [[Rugby_league_county_cups|County Cup]] Final during the [[1934–35_Northern_Rugby_Football_League_season|1934–35 season]] at [[Station_Road,_Swinton|Station Road, Swinton]] on Saturday 20 October 1934, the 15–7 victory over [[Wigan_Warriors|Wigan]] in the [[1935_Lancashire_Cup|1935]] [[RFL_Lancashire_Cup|Lancashire]] [[Rugby_league_county_cups|County Cup]] Final during the [[1935–36_Northern_Rugby_Football_League_season|1935–36 season]] at [[Wilderspool_Stadium|Wilderspool Stadium]], [[Warrington|Warrington]] on Saturday 19 October 1935, the 5–2 victory over [[Wigan_Warriors|Wigan]] in the [[1936_Lancashire_Cup|1936]] [[RFL_Lancashire_Cup|Lancashire]] [[Rugby_league_county_cups|County Cup]] Final during the [[1936–37_Northern_Rugby_Football_League_season|1936–37 season]] at [[Wilderspool_Stadium|Wilderspool Stadium]], [[Warrington|Warrington]] on Saturday 17 October 1936, and he played [[Fullback_(rugby_league)|fullback]] in the 7–10 defeat by [[Wigan_Warriors|Wigan]] in the [[1938_Lancashire_Cup|1938]] [[RFL_Lancashire_Cup|Lancashire]] [[Rugby_league_county_cups|County Cup]] Final during the [[1938–39_Northern_Rugby_Football_League_season|1938–39 season]] at [[Station_Road,_Swinton|Station Road, Swinton]] on Saturday 22 October 1938.\n[[Jim Sullivan (Welsh rugby league)_sentence_2]]: A a right-footed [[Toe_punt|toe-end]] style (rather than [[Round_the_corner_kicking|round the corner]] style) [[Goal_(sport)#Rugby_football|goal]]-kicking [[Fullback_(rugby_league)|fullback]], he scored 4,883 points in a career that spanned 25 years with Wigan, and still holds several records with the club today.\n[[Jim Sullivan (Welsh rugby league)_sentence_8]]: In December 1920, 26-days after his seventeenth birthday, he played for the [[Barbarian_F.C.|Barbarians]] in a match against [[Newport_RFC|Newport]], becoming the youngest player to represent the team.\n[[Jim Sullivan (Welsh rugby league)_sentence_17]]: Jim Sullivan played [[Fullback_(rugby_league)|fullback]], and scored 4 conversions in Wigan's 20–2 victory over [[Leigh_Centurions|Leigh]] in the [[1922–23_Lancashire_Cup|1922–23 Lancashire Cup]] Final at [[The_Willows,_Salford|the Willows]] in Weaste, [[County_Borough_of_Salford|Salford]] on Saturday 25 November 1922, played [[Fullback_(rugby_league)|fullback]], and scored a conversion in the 11–15 defeat by Swinton in the [[1925–26_Lancashire_Cup|1925–26 Lancashire Cup]] Final at The Cliff, Broughton on Wednesday 9 December 1925, played [[Fullback_(rugby_league)|fullback]], and scored a conversion in the 2–5 defeat by Swinton in the [[1927–28_Lancashire_Cup|1927–28 Lancashire Cup]] Final at [[Watersheddings|Watersheddings]], [[County_Borough_of_Oldham|Oldham]] on Saturday 19 November 1927, played [[Fullback_(rugby_league)|fullback]], and scored a conversion in the 5–4 victory over Widnes in the [[1928–29_Lancashire_Cup|1928–29 Lancashire Cup]] Final the Willows, Salford, on Saturday 24 November 1928, played [[Fullback_(rugby_league)|fullback]], and scored 2-conversions in the 12–21 defeat by Salford in the [[1934–35_Lancashire_Cup|1934–35 Lancashire Cup]] Final at [[Station_Road,_Swinton|Station Road]], [[Swinton,_Greater_Manchester|Swinton]], on Saturday 20 October 1934, played [[Fullback_(rugby_league)|fullback]], and scored 2-conversions in the 7–15 defeat by Salford in the [[1935–36_Lancashire_Cup|1935–36 Lancashire Cup]] Final at Wilderspool Stadium, Warrington on Saturday 19 October 1935, played [[Fullback_(rugby_league)|fullback]], and scored a conversion in the 2–5 defeat by Salford in the [[1936–37_Lancashire_Cup|1936–37 Lancashire Cup]] Final at Wilderspool, Warrington on Saturday 17 October 1936, and played [[Fullback_(rugby_league)|fullback]] and scored 5-conversions in the 10–7 victory over Salford in the [[1938–39_Lancashire_Cup|1938–39 Lancashire Cup]] Final during the [[1938–39_Northern_Rugby_Football_League_season|1938–39 season]] at Station Road, Swinton on Saturday 22 October 1938.\n[[Jim Sullivan (Welsh rugby league)_sentence_29]]: He had won three league Championships, two Challenge Cups and three Lancashire Cups.\n</information>\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Jim Sullivan (Welsh rugby league)_sentence_17",
"Harold Osbaldestin_sentence_6"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 8182,
"split": "train_offline"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "78.603",
"value": "undefined"
},
{
"operation": "search",
"time": "126.831",
"value": "1938-39 Lancashire Cup"
},
{
"operation": "Now on",
"time": "132.346",
"value": "1938-39 Lancashire Cup"
},
{
"operation": "Highlighting",
"time": "140.306",
"value": "1938–39 Lancashire Cup_cell_1_3_0"
},
{
"operation": "Highlighting",
"time": "140.308",
"value": "1938–39 Lancashire Cup_title"
},
{
"operation": "Highlighting",
"time": "140.309",
"value": "1938–39 Lancashire Cup_section_1"
},
{
"operation": "Highlighting",
"time": "140.31",
"value": "1938–39 Lancashire Cup_section_5"
},
{
"operation": "Highlighting",
"time": "140.31",
"value": "1938–39 Lancashire Cup_section_6"
},
{
"operation": "Highlighting",
"time": "140.314",
"value": "1938–39 Lancashire Cup_header_cell_1_0_0"
},
{
"operation": "Highlighting",
"time": "147.485",
"value": "1938–39 Lancashire Cup_cell_1_3_2"
},
{
"operation": "Highlighting",
"time": "147.489",
"value": "1938–39 Lancashire Cup_header_cell_1_0_2"
},
{
"operation": "hyperlink",
"time": "164.86",
"value": "Jim Sullivan"
},
{
"operation": "Now on",
"time": "166.849",
"value": "Jim Sullivan (Welsh rugby league)"
},
{
"operation": "Highlighting",
"time": "175.908",
"value": "Jim Sullivan (Welsh rugby league)_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "179.346",
"value": "Jim Sullivan (Welsh rugby league)_sentence_2"
},
{
"operation": "hyperlink",
"time": "201.526",
"value": "1938–39 Lancashire Cup"
},
{
"operation": "Now on",
"time": "202.462",
"value": "1938–39 Lancashire Cup"
},
{
"operation": "hyperlink",
"time": "210.588",
"value": "1938–39 Northern Rugby Football League season"
},
{
"operation": "Now on",
"time": "211.442",
"value": "1938–39 Northern Rugby Football League season"
},
{
"operation": "back-button-clicked",
"time": "216.943",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "217.062",
"value": "1938–39 Lancashire Cup"
},
{
"operation": "hyperlink",
"time": "228.449",
"value": "Jim Sullivan"
},
{
"operation": "Now on",
"time": "228.87",
"value": "Jim Sullivan (Welsh rugby league)"
},
{
"operation": "Highlighting",
"time": "243.513",
"value": "Jim Sullivan (Welsh rugby league)_sentence_17"
},
{
"operation": "back-button-clicked",
"time": "257.707",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "257.845",
"value": "1938–39 Lancashire Cup"
},
{
"operation": "hyperlink",
"time": "262.697",
"value": "Harold Osbaldestin"
},
{
"operation": "Now on",
"time": "263.297",
"value": "Harold Osbaldestin"
},
{
"operation": "Highlighting",
"time": "278.909",
"value": "Harold Osbaldestin_sentence_6"
},
{
"operation": "finish",
"time": "317.503",
"value": "finish"
}
]
}
|
61200
|
The 1999 Anguillian general election was held on the fourth of March and the opposition Anguilla National Alliance won three seats.
|
{
"evidence": [
"1999 Anguillian general election_sentence_0",
"1999 Anguillian general election_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information above.\n\n<think>\n...\n</think>\n- Use for every piece of reasoning.\n- During reasoning, you must verify the claim step by step based on the given information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer>\n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\nVerify the claim:\nThe 1999 Anguillian general election was held on the fourth of March and the opposition Anguilla National Alliance won three seats.\n<information>\n[[1999 Anguillian general election_sentence_0]]: General elections were held in [[Anguilla|Anguilla]] on 4 March 1999.\n[[1999 Anguillian general election_sentence_1]]: The result was a victory for the ruling coalition of the Anguilla United Party and the [[Anguilla_Democratic_Party|Anguilla Democratic Party]], each winning 2 seats.\n[[1999 Anguillian general election_sentence_2]]: The opposition [[Anguilla_National_Alliance|Anguilla National Alliance]] won 3 seats.\n[[1999 Anguillian general election_sentence_3]]: However, the coalition government lost its majority in May 1999 upon the resignation of [[Victor_Banks|Victor Banks]], the Finance Minister and leader of the [[Anguilla_Democratic_Party|Anguilla Democratic Party]].\n[[1999 Anguillian general election_sentence_4]]: Due to the collapse of the ruling coalition, [[2000_Anguillan_general_election|fresh elections]] were held in March 2000.\n</information>\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"1999 Anguillian general election_sentence_0",
"1999 Anguillian general election_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 8183,
"split": "train_offline"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.636",
"value": "?search="
},
{
"operation": "search",
"time": "10.159",
"value": "1999 Anguillian general election"
},
{
"operation": "Now on",
"time": "11.325",
"value": "1999 Anguillian general election"
},
{
"operation": "Highlighting",
"time": "41.409",
"value": "1999 Anguillian general election_sentence_0"
},
{
"operation": "Highlighting",
"time": "45.855",
"value": "1999 Anguillian general election_sentence_2"
},
{
"operation": "finish",
"time": "96.499",
"value": "finish"
}
]
}
|
exfever_7771
|
Bernard Sanders is an American politician and activist who has served as the junior United States senator from a state in the New England region of the United States.
|
{
"evidence": [
"Bernie Sanders",
"Vermont"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information above.\n\n<think>\n...\n</think>\n- Use for every piece of reasoning.\n- During reasoning, you must verify the claim step by step based on the given information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer>\n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\nVerify the claim:\nBernard Sanders is an American politician and activist who has served as the junior United States senator from a state in the New England region of the United States.\n<information>\n[[Gabriel Manigault]]: Gabriel Manigault (March 17, 1758 – November 4, 1809) was an American architect. He was born in Charleston, South Carolina, went to study in Geneva, Switzerland, and London, and came back to Charleston after the American Revolutionary War. He died in Philadelphia, Pennsylvania. His great-grandfather was Pierre Manigault (1664-1729), a French Huguenot who was born in La Rochelle, France and settled in Charleston. His father was Peter Manigault (1731-1773), the wealthiest person in British North America in 1770.\n\n[[Trunk Monkey]]: Trunk Monkey is a concept originally created in January, 2000 by Sean Sosik-Hamor for the automotive group of Subaru enthusiasts in the New England Subaru Impreza Club, or NESIC. The original concept of a \"Trunk Monkey\" was a small stuffed animal placed in the trunk of a vehicle which came alive and jumped from side to side to help distribute weight during \"spirited driving maneuvers\".\n\n[[John Lumsden (footballer)]]: John David Lumsden (15 December 1960 – 22 April 2016) was a Scottish footballer who played in the Football League for Stoke City.\n\n[[Bernie Sanders]]: Bernard Sanders (born September 8, 1941) is an American politician who has been the junior United States Senator from Vermont since 2007. Sanders is the longest serving independent in U.S. congressional history. Since his election to the House of Representatives in 1990, he has caucused with the Democratic Party, which has entitled him to congressional committee assignments and at times given Democrats a majority. Sanders became the ranking minority member on the Senate Budget Committee in January 2015; he had previously been chair of the Senate Veterans' Affairs Committee for two years.\n\n[[Jamuna Rani]]: K. Jamuna Rani (born 17 May 1938 in Andhra Pradesh, India) is a prolific Indian Playback singer who has sung over 6,000 songs in Telugu, Tamil, Kannada, Malayalam and Sinhala languages.\n\n[[Vermont]]: Vermont ( ) is a state in the New England region of the northeastern United States. It borders the other U.S. states of Massachusetts to the south, New Hampshire to the east, New York to the west, and the Canadian province of Quebec to the north. Lake Champlain forms half of Vermont's western border with the state of New York and the Green Mountains run north-south the length of the state.\n\n[[Jimmy Cornell]]: Jimmy Cornell is a Romanian-born British yachtsman, bestselling author and the founder of the World Cruising Club.\n</information>\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Bernie Sanders",
"Vermont"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 8184,
"split": "train_offline"
}
|
{
"annotator_operations": []
}
|
exfever_4310
|
Saddam Hussein was an Iraqi politician who served as the country's fifth president and was a leading role in the revolutionary Arab Socialist Ba'ath Party, which supported Ba'athism, a mix of Arab socialism and nationalist ideology that affirms the Arabs as a nation and advocates Arab unity.
|
{
"evidence": [
"Saddam Hussein",
"Arab nationalism"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information above.\n\n<think>\n...\n</think>\n- Use for every piece of reasoning.\n- During reasoning, you must verify the claim step by step based on the given information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer>\n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\nVerify the claim:\nSaddam Hussein was an Iraqi politician who served as the country's fifth president and was a leading role in the revolutionary Arab Socialist Ba'ath Party, which supported Ba'athism, a mix of Arab socialism and nationalist ideology that affirms the Arabs as a nation and advocates Arab unity.\n<information>\n[[Venture Community F.C.]]: Venture Community Football Club (Welsh: Clwb Pel Droed Y Fentur) is an association football club based in Wrexham, Wales, formed in 1994.\n\n[[Arab nationalism]]: Arab nationalism (Arabic: القومية العربية \"al-Qawmiyya al-`arabiyya\") is a nationalist ideology that asserts that the Arabs are a nation and promotes the unity of Arab people, celebrating the glories of Arab civilization, the language and literature of the Arabs, calling for rejuvenation and political union in the Arab world. Its central premise is that the peoples of the Arab world, from the Atlantic Ocean to the Indian Ocean, constitute one nation bound together by common ethnic, linguistic, cultural, historical, identical, nationalist, geographical and political. One of the primary goals of Arab nationalism is the end of Western influence in the Arab world, seen as a \"nemesis\" of Arab strength, and the removal of those Arab governments considered to be dependent upon Western power. It rose to prominence with the weakening and defeat of the Ottoman Empire in the early 20th century and declined after the defeat of the Arab armies in the Six-Day War.\n\n[[David Hanson (computer scientist)]]: David R. Hanson is a Software Engineer who worked at Google until he retired in January 2012, working in programming languages, compilers, software tools, and programming environments. Before joining Google, he was with Microsoft Research, Princeton, University of Arizona, and Yale. He has written many journal and conference papers and two books: \"A Retargetable C Compiler: Design and Implementation\" (ISBN , with Chris Fraser), which describes lcc, a widely used compiler for Standard C, and \"C Interfaces and Implementations: Techniques for Creating Reusable Software\" (ISBN ).\n\n[[Sunakothi]]: Sunakothi is a small town located about 4 km south of the main Lalitpur town in Lalitpur District. According to 2011 Nepal census, Sunakothi has a population of 10,092 living in 2397 individual households. Most of the people living in Sunakothi are Newar. The offshoot road near the Satdobato segment of the Ring Road leads to Sunakothi. This road ultimately leads to Lele and is referred to as Laxmi Prasad Devkota Marga (Satdobato - Tika Bahirab).Thecho is situated on the southern part of Sunakothi, Bungamati and Bhaisipati on west, Dhapakhel on east and Khumaltar and Nakhipot on North.\n\n[[Saint-Pierre-lès-Bitry]]: Saint-Pierre-lès-Bitry is a commune in the Oise department in northern France.\n\n[[Saddam Hussein]]: Saddam Hussein Abd al-Majid al-Tikriti ( ; Arabic: صدام حسين عبد المجيد التكريتي \"Ṣaddām Ḥusayn ʿAbd al-Maǧīd al-Tikrītī \"; 28 April 1937 – 30 December 2006) was the fifth President of Iraq, serving in this capacity from 16 July 1979 until 9 April 2003. A leading member of the revolutionary Arab Socialist Ba'ath Party, and later, the Baghdad-based Ba'ath Party and its regional organization the Iraqi Ba'ath Party—which espoused Ba'athism, a mix of Arab nationalism and socialism—Saddam played a key role in the 1968 coup (later referred to as the 17 July Revolution) that brought the party to power in Iraq.\n</information>\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Saddam Hussein",
"Arab nationalism"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 8185,
"split": "train_offline"
}
|
{
"annotator_operations": []
}
|
exfever_2747
|
21 Jump Street is a 2012 American buddy cop action comedy film adapted from the television series of the same name by an American television producer and writer.
|
{
"evidence": [
"21 Jump Street (film)",
"Patrick Hasburgh"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information above.\n\n<think>\n...\n</think>\n- Use for every piece of reasoning.\n- During reasoning, you must verify the claim step by step based on the given information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer>\n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\nVerify the claim:\n21 Jump Street is a 2012 American buddy cop action comedy film adapted from the television series of the same name by an American television producer and writer.\n<information>\n[[2007 IndyCar Series]]: The 2007 IRL IndyCar Series began with a night race on Saturday March 24 at Homestead-Miami Speedway. The season's premiere event, the 91st Indianapolis 500 was held May 27. The season finale was held at Chicagoland Speedway on September 9. Dario Franchitti, who won four races during the season, including the Indy 500, clinched the 2007 IndyCar Series Championship on the final lap of the final race, by winning the race after points leader Scott Dixon ran out of fuel while leading with less than ⅓ of a lap to go.\n\n[[Liechtenstein at the Paralympics]]: Liechtenstein made its Paralympic Games début at the 1984 Summer Paralympics in Stoke Mandeville and New York City, entering just one athlete (Iris Schaedler) in athletics. The country competed again at the 1988 and 1992 Summer Games, and at the 1992 and 1994 Winter Games. It was then absent from the Paralympics until the 2004 Summer Games in Athens, where it sent just one table tennis player, Peter Frommelt. Liechtenstein was absent from the 2006 and 2010 Winter Games and the 2008 Summer Games.\n\n[[Gerrymandering in the United States]]: Gerrymandering in the United States has been practiced since the founding of the country to strengthen the power of particular political interests within legislative bodies. Partisan gerrymandering is commonly used to increase the power of a political party. In some instances, political parties collude to protect incumbents by engaging in bipartisan gerrymandering. After racial minorities were enfranchised, some jurisdictions engaged in racial gerrymandering to weaken the political power of racial minority voters, while others engaged in racial gerrymandering to strengthen the power of minority voters. Throughout the 20th century, courts have grappled with the legality of these types of gerrymandering and have devised different standards for the different types of gerrymandering. Various legal and political remedies have emerged to prevent gerrymandering, including court-ordered redistricting plans, redistricting commissions, and alternative voting systems that do not depend on drawing boundaries for single-member electoral districts.\n\n[[Patrick Hasburgh]]: Patrick Hasburgh is an American television producer and writer.\n\n[[21 Jump Street (film)]]: 21 Jump Street is a 2012 American action comedy film directed by Phil Lord and Christopher Miller, written by Michael Bacall starring Jonah Hill and Channing Tatum. An adaptation of the 1987 television series of the same name by Stephen J. Cannell and Patrick Hasburgh, the film follows two police officers who are forced to relive high school when they are assigned to go undercover as high school students to prevent the outbreak of a new synthetic drug and arrest its supplier.\n\n[[Labriola v. Pollard Group, Inc.]]: Labriola v. Pollard Group, Inc., 152 Wash.2d 828 (2004), was a case decided by the Supreme Court of Washington that invalidated a modification of an employment contract for an at-will employee which would have added a non-competition agreement because the modification lacked independent consideration.\n\n[[Crested servaline genet]]: The crested servaline genet (\"Genetta cristata\"), also known as the crested genet, is a genet species endemic to Nigeria and Cameroon. As the population has declined due to loss of habitat, it is listed as Vulnerable on the IUCN Red List. It was first recorded in the Mamfe Division in Cameroon and initially considered a subspecies of the servaline genet (\"Genetta servalina\"). But now it is regarded as a distinct species.\n</information>\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"21 Jump Street (film)",
"Patrick Hasburgh"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 8186,
"split": "train_offline"
}
|
{
"annotator_operations": []
}
|
exfever_8667
|
Nine Inch Nail's dark, even sadistic themes, stood in stark contrast to things like the Spice Girls and Britney Spears, who were also popular at the time.
|
{
"evidence": [
"Year Zero (album)",
"Concept album"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information above.\n\n<think>\n...\n</think>\n- Use for every piece of reasoning.\n- During reasoning, you must verify the claim step by step based on the given information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer>\n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\nVerify the claim:\nNine Inch Nail's dark, even sadistic themes, stood in stark contrast to things like the Spice Girls and Britney Spears, who were also popular at the time.\n<information>\n[[Concept album]]: A concept album is an album in which its tracks hold a larger purpose or meaning collectively than they do individually. This is typically achieved through a single central narrative or theme, which can be instrumental, compositional, or lyrical. Sometimes the term is referenced to albums considered to be of \"uniform excellence\" rather than an LP with an explicit musical or lyrical motif. The exact criterion for a \"concept album\" varies, with no discernible consensus.\n\n[[Joe Crowley (television presenter)]]: Joe Crowley is an English television presenter and broadcast journalist, best known for presenting and reporting on \"The One Show\" and \"Countryfile\" as well as co-presenting the factual BBC One series \"Holiday Hit Squad\" since 2013, alongside Angela Rippon and Helen Skelton.\n\n[[Microregion]]: Microregion is a designation for territorial entities.\n\n[[1997–98 Liga Artzit]]: The 1997–98 Liga Artzit season saw Maccabi Jaffa win the title and earn promotion to the top division. Tzafririm Holon were also promoted.\n\n[[Rosstown, Indiana]]: Rosstown is an unincorporated community in Wayne Township, Bartholomew County, Indiana.\n\n[[Olympus C-765 Ultra Zoom]]: The Olympus Camedia 765 Ultra Zoom is a 4.0 megapixel camera with a 10x optical zoom manufactured by Olympus starting in 2004. It is powered by a lithium-ion battery.\n\n[[Year Zero (album)]]: Year Zero is the fifth studio album by American industrial rock band Nine Inch Nails, released on April 17, 2007 by Interscope Records. Frontman Trent Reznor wrote the album's music and lyrics while touring in support of the band's previous album \"With Teeth\" (2005). In contrast to the introspective style of songwriting Reznor used on previous albums, \"Year Zero\" is a concept album that criticizes contemporary policies of the United States government by presenting a dystopian vision of the year 2022.\n</information>\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Year Zero (album)",
"Concept album"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 8187,
"split": "train_offline"
}
|
{
"annotator_operations": []
}
|
exfever_6257
|
Neymar is a Brazilian professional footballer who plays for Paris Saint-Germain as a player who plays nearest to the opposing team's goal, in a team sport played with a spherical ball between two teams of 14 players.
|
{
"evidence": [
"Neymar",
"Forward (association football)",
"Association football"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information above.\n\n<think>\n...\n</think>\n- Use for every piece of reasoning.\n- During reasoning, you must verify the claim step by step based on the given information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer>\n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\nVerify the claim:\nNeymar is a Brazilian professional footballer who plays for Paris Saint-Germain as a player who plays nearest to the opposing team's goal, in a team sport played with a spherical ball between two teams of 14 players.\n<information>\n[[Flip, Flop and Fly]]: \"Flip, Flop and Fly\" is a jump blues-style song recorded by Big Joe Turner in 1955. Called a \"prototypical rocker,\" the song was a hit reaching #2 in \"Billboard\" magazine's R&B chart. \"Flip, Flop and Fly\" has been recorded by a variety of artists, including early rock and roll performers such as Elvis Presley.\n\n[[Association football]]: Association football, more commonly known as football or soccer, is a team sport played between two teams of eleven players with a spherical ball. It is played by 250 million players in over 200 countries and dependencies, making it the world's most popular sport. The game is played on a rectangular field with a goal at each end. The object of the game is to score by getting the ball into the opposing goal.\n\n[[Forward (association football)]]: Forwards are the players on an association football team who play nearest to the opposing team's goal, and are therefore most responsible for scoring goals.\n\n[[Alpha Kappa Nu]]: Alpha Kappa Nu (ΑΚΝ) is the first documented African-American collegiate fraternal organization in the United States. Founded in 1903 at Indiana University and lasting until around 1905, it had a membership of 10, which included all the colored students at the school. A second chapter was set to be established at Wilberforce University. Wilberforce University was where the fraternity Gamma Phi was established in 1905.\n\n[[Wrestling at the 2014 Asian Games – Men's freestyle 57 kg]]: Wrestling at the 2014 Asian Games – Men's freestyle 57 kg\n\n[[Neymar]]: Neymar da Silva Santos Júnior (] ; born 5 February 1992), commonly known as Neymar or Neymar Jr., is a Brazilian professional footballer who plays as a forward for French club Paris Saint-Germain and the Brazil national team.\n</information>\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Neymar",
"Forward (association football)",
"Association football"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 8188,
"split": "train_offline"
}
|
{
"annotator_operations": []
}
|
29978
|
After the concurrency Highway 2 continues northward, meeting US-270 at Wilburton; SH-2 then passes Robbers Cave State Park and the eastern edge of the Sansbois Mountains before reaching SH-31 east of Quinton.
|
{
"evidence": [
"Oklahoma State Highway 2_sentence_14",
"Oklahoma State Highway 2_sentence_15"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information above.\n\n<think>\n...\n</think>\n- Use for every piece of reasoning.\n- During reasoning, you must verify the claim step by step based on the given information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer>\n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\nVerify the claim:\nAfter the concurrency Highway 2 continues northward, meeting US-270 at Wilburton; SH-2 then passes Robbers Cave State Park and the eastern edge of the Sansbois Mountains before reaching SH-31 east of Quinton.\n<information>\n[[Oklahoma State Highway 2_sentence_0]]: \"OK 2\" redirects here.\n[[Oklahoma State Highway 2_sentence_1]]: OK 2 may also refer to [[Oklahoma's_2nd_congressional_district|Oklahoma's 2nd congressional district]].\n[[Oklahoma State Highway 2_sentence_2]]: State Highway 2, abbreviated SH-2 or OK-2, is a designation for two distinct highways maintained by the [[United_States|U.S.]]\n[[Oklahoma State Highway 2_sentence_11]]: Immediately after this, SH-2 meets [[State_Highway_43_(Oklahoma)|SH-43]]'s eastern terminus southeast of [[Sardis_Lake_(Oklahoma)|Sardis Lake]].\n[[Oklahoma State Highway 2_sentence_14]]: After the concurrency Highway 2 continues northward, meeting [[U.S._Highway_270_(Oklahoma)|US-270]] at [[Wilburton,_Oklahoma|Wilburton]].\n[[Oklahoma State Highway 2_sentence_15]]: SH-2 then passes Robbers Cave State Park and the eastern edge of the Sansbois Mountains before reaching [[State_Highway_31_(Oklahoma)|SH-31]] east of [[Quinton,_Oklahoma|Quinton]], and the two form a six-mile (10 km) concurrency until [[Kinta,_Oklahoma|Kinta]].\n[[Oklahoma State Highway 2_sentence_18]]: 10 miles (16 km) later, SH-2 meets [[U.S._Highway_266_(Oklahoma)|US-266]], which it will overlap until its end.\n[[Oklahoma State Highway 2_sentence_31]]: The old route on Bull Creek Mountain was decommissioned and is no longer in use.\n</information>\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Oklahoma State Highway 2_sentence_14",
"Oklahoma State Highway 2_sentence_15"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 8189,
"split": "train_offline"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.774",
"value": "?search="
},
{
"operation": "search",
"time": "30.339",
"value": "SH-2"
},
{
"operation": "Now on",
"time": "31.299",
"value": "Kaman SH-2 Seasprite"
},
{
"operation": "hyperlink",
"time": "37.01",
"value": "SH2 (disambiguation)"
},
{
"operation": "Now on",
"time": "37.428",
"value": "SH2"
},
{
"operation": "hyperlink",
"time": "47.121",
"value": "List of highways numbered 2"
},
{
"operation": "Now on",
"time": "47.999",
"value": "List of highways numbered 2"
},
{
"operation": "hyperlink",
"time": "85.51",
"value": "Oklahoma State Highway 2"
},
{
"operation": "Now on",
"time": "86.072",
"value": "Oklahoma State Highway 2"
},
{
"operation": "Highlighting",
"time": "98.454",
"value": "Oklahoma State Highway 2_sentence_14"
},
{
"operation": "Highlighting",
"time": "103.549",
"value": "Oklahoma State Highway 2_sentence_15"
},
{
"operation": "finish",
"time": "127.942",
"value": "finish"
}
]
}
|
6974
|
In Ludruk, a simple and straightforward comedy, actors would almost always use the Surabaya dialect of Javanese language (language of the Javanese people from the central and eastern parts of the island of Java).
|
{
"evidence": [
"Ludruk_sentence_5",
"Ludruk_sentence_4",
"Javanese language_sentence_3"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information above.\n\n<think>\n...\n</think>\n- Use for every piece of reasoning.\n- During reasoning, you must verify the claim step by step based on the given information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer>\n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\nVerify the claim:\nIn Ludruk, a simple and straightforward comedy, actors would almost always use the Surabaya dialect of Javanese language (language of the Javanese people from the central and eastern parts of the island of Java).\n<information>\n[[Javanese language_sentence_3]]: Javanese ([[English|/dʒɑːvəˈniːz/]]; ꦧꦱꦗꦮ, Basa Jawa, باسا جاوا, Javanese pronunciation: [[Help:IPA|[bɔsɔ d͡ʒɔwɔ]]; colloquially known as ꦕꦫꦗꦮ, Cara Jawa, Javanese pronunciation: [[Help:IPA|[t͡ʃɔrɔ d͡ʒɔwɔ]]) is the language of the [[Javanese_people|Javanese people]] from the central and eastern parts of the island of [[Java|Java]], in Indonesia.\n[[Javanese language_sentence_19]]: In general, the history of the Javanese language can be divided into two distinct phases: 1) Old Javanese and 2) New Javanese.\n[[Javanese language_sentence_172]]: The vocabulary of standard Javanese is enriched by dialectal words.\n[[Javanese language_sentence_218]]: There is, however, no longer a daily newspaper in Javanese.\n[[Ludruk_sentence_0]]: Ludruk is one of the theatrical genres in [[East_Java|East Java]].\n[[Ludruk_sentence_1]]: It is a form of traditional performance presented by a troupe of actors (or comedians) on a stage, re-telling the life stories of everyday people and their struggles.\n[[Ludruk_sentence_4]]: The actors would almost always use the [[Surabaya|Surabaya]] [[Dialect|dialect]] of [[Javanese_language|Javanese language]], although sometimes there can be occasional guest stars from other areas, such as [[Jombang_Regency|Jombang]], [[Malang|Malang]], [[Madura|Madura]], [[Madiun|Madiun]], who each would use their own dialect.\n[[Ludruk_sentence_5]]: Despite the fact that many different dialects are employed in one performance, as a whole, ludruk is a simple and straightforward comedy, making ludruk easily understood by everyone.\n[[Ludruk_sentence_7]]: A typical ludruk performance begins with a performance of Remo dance and followed by a portrayal Pak Sakera, a legendary 19th century [[Madurese_people|Madurese]] hero.\n[[Ludruk_sentence_9]]: Both genres focus on delivering a particular moral message, but ludruk tells stories about everyday life, and mostly that of the underprivileged.\n</information>\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Ludruk_sentence_5",
"Ludruk_sentence_4",
"Javanese language_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 8190,
"split": "train_offline"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.05",
"value": "?search="
},
{
"operation": "search",
"time": "16.387",
"value": "Ludruk"
},
{
"operation": "Now on",
"time": "16.971",
"value": "Ludruk"
},
{
"operation": "Highlighting",
"time": "131.845",
"value": "Ludruk_sentence_3"
},
{
"operation": "Highlighting",
"time": "136.501",
"value": "Ludruk_sentence_5"
},
{
"operation": "Highlighting deleted",
"time": "137.158",
"value": "Ludruk_sentence_3"
},
{
"operation": "Highlighting",
"time": "140.798",
"value": "Ludruk_sentence_4"
},
{
"operation": "hyperlink",
"time": "149.398",
"value": "Javanese language"
},
{
"operation": "Now on",
"time": "151.907",
"value": "Javanese language"
},
{
"operation": "Highlighting",
"time": "157.902",
"value": "Javanese language_sentence_3"
},
{
"operation": "finish",
"time": "166.25",
"value": "finish"
}
]
}
|
74026
|
Arts editor for The Catholic Herald, Miguel Cullen, gave up writing to poetry focus on journalism.
|
{
"evidence": [
"Miguel Cullen_sentence_0"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information above.\n\n<think>\n...\n</think>\n- Use for every piece of reasoning.\n- During reasoning, you must verify the claim step by step based on the given information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer>\n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\nVerify the claim:\nArts editor for The Catholic Herald, Miguel Cullen, gave up writing to poetry focus on journalism.\n<information>\n[[Miguel Cullen_sentence_0]]: Miguel Cullen is a [[United_Kingdom|British]] [[Poet|poet]] and [[Journalist|journalist]] who lives in [[London|London]].\n[[Miguel Cullen_sentence_12]]: Poet Steven J Fowler said of it: \"Lyrical, voluminously expressive, beautiful in their knotted, winding intensity – Miguel Cullen's poems are intricate, funny for everyone but you, unpleasantly bright and brilliant.\"\n[[Miguel Cullen_sentence_13]]: Cullen's poetry has been described as imagistic, inflected with urban vernacular, and preoccupied with iconic film and urban music references to create a delicate, interconnected world.\n[[Miguel Cullen_sentence_14]]: The critic [[Ian_Thomson_(writer)|Ian Thomson (writer)]] compared him to [[Ezra_Pound|Ezra Pound]], citing his \"allusions from Greek mythology (colliding) with sound system culture (and) pavement pounding street demotic\".\n</information>\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Miguel Cullen_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 8191,
"split": "train_offline"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.754",
"value": "?search="
},
{
"operation": "search",
"time": "16.825",
"value": "Miguel Cullen"
},
{
"operation": "Now on",
"time": "17.478",
"value": "Miguel Cullen"
},
{
"operation": "Highlighting",
"time": "43.451",
"value": "Miguel Cullen_sentence_0"
},
{
"operation": "finish",
"time": "58.651",
"value": "finish"
}
]
}
|
10041
|
Hog Island Light is wrongly believed to have been first lit in 1853 and is located at Hog Island SE of Exmore, Virginia (population 1,460).
|
{
"evidence": [
"Hog Island Light_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:\nHog Island Light is wrongly believed to have been first lit in 1853 and is located at Hog Island SE of Exmore, Virginia (population 1,460).\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Hog Island Light_sentence_6"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 0,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.551",
"value": "?search="
},
{
"operation": "search",
"time": "12.285",
"value": "Hog Island Light"
},
{
"operation": "Now on",
"time": "12.952",
"value": "Hog Island Light"
},
{
"operation": "Highlighting",
"time": "29.289",
"value": "Hog Island Light_header_cell_0_2_0"
},
{
"operation": "Highlighting",
"time": "29.702",
"value": "Hog Island Light_cell_0_2_1"
},
{
"operation": "Highlighting",
"time": "29.706",
"value": "Hog Island Light_header_cell_0_2_0"
},
{
"operation": "Highlighting",
"time": "57.611",
"value": "Hog Island Light_sentence_6"
},
{
"operation": "Highlighting deleted",
"time": "58.932",
"value": "Hog Island Light_cell_0_2_1"
},
{
"operation": "Highlighting deleted",
"time": "58.935",
"value": "Hog Island Light_header_cell_0_2_0"
},
{
"operation": "finish",
"time": "64.978",
"value": "finish"
}
]
}
|
exfever_8865
|
The Fate of the Furious is an Filipino action film starring an American actor who also appear in a supporting role in an American epic war film directed by Steven Spielberg.
|
{
"evidence": [
"The Fate of the Furious",
"Vin Diesel",
"Saving Private Ryan"
],
"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 Fate of the Furious is an Filipino action film starring an American actor who also appear in a supporting role in an American epic war film directed by Steven Spielberg.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Fate of the Furious",
"Vin Diesel",
"Saving Private Ryan"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 1,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6938
|
The Host is a 2013 American film based on an American novel published in 2008 by a novelist and film producer.
|
{
"evidence": [
"The Host (2013 film)",
"Stephenie Meyer"
],
"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 Host is a 2013 American film based on an American novel published in 2008 by a novelist and film producer.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"The Host (2013 film)",
"Stephenie Meyer"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 2,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5186
|
The English Civil War was a series of civil wars and political machinations between Parliamentarians and Cavalier, a term of abuse for the wealthier Royalist supporters of King Charles I and his son, the King of Scotland, from 1649 until 1654.
|
{
"evidence": [
"English Civil War",
"Cavalier",
"Charles II of England"
],
"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 English Civil War was a series of civil wars and political machinations between Parliamentarians and Cavalier, a term of abuse for the wealthier Royalist supporters of King Charles I and his son, the King of Scotland, from 1649 until 1654.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"English Civil War",
"Cavalier",
"Charles II of England"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 3,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
82613
|
Geghuni Chitchian is an Armenian composer and singer who became an Honoured Representative of the Arts in 1980.
|
{
"evidence": [
"Geghuni Chitchian_sentence_0",
"Geghuni Chitchian_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:\nGeghuni Chitchian is an Armenian composer and singer who became an Honoured Representative of the Arts in 1980.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Geghuni Chitchian_sentence_0",
"Geghuni Chitchian_sentence_5"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 4,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.652",
"value": "?search="
},
{
"operation": "search",
"time": "13.226",
"value": "Geghuni Chitchian"
},
{
"operation": "Now on",
"time": "13.835",
"value": "Geghuni Chitchian"
},
{
"operation": "Highlighting",
"time": "18.876",
"value": "Geghuni Chitchian_sentence_0"
},
{
"operation": "Page search",
"time": "30.184",
"value": "sing"
},
{
"operation": "Highlighting",
"time": "70.281",
"value": "Geghuni Chitchian_sentence_5"
},
{
"operation": "finish",
"time": "234.129",
"value": "finish"
}
]
}
|
19027
|
Anatole Bailly was interested in history, authored many textbooks but did not join any society.
|
{
"evidence": [
"Anatole Bailly_sentence_11"
],
"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:\nAnatole Bailly was interested in history, authored many textbooks but did not join any society.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Anatole Bailly_sentence_11"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 5,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.201",
"value": "?search="
},
{
"operation": "search",
"time": "12.081",
"value": "Anatole Bailly"
},
{
"operation": "Now on",
"time": "13.119",
"value": "Anatole Bailly"
},
{
"operation": "Highlighting",
"time": "108.901",
"value": "Anatole Bailly_sentence_10"
},
{
"operation": "Highlighting",
"time": "114.468",
"value": "Anatole Bailly_sentence_11"
},
{
"operation": "Highlighting deleted",
"time": "115.195",
"value": "Anatole Bailly_sentence_10"
},
{
"operation": "finish",
"time": "125.004",
"value": "finish"
}
]
}
|
77707
|
Reece Mastin peaked at No. 2 in the Austrian music charts; then peaked at No. 1 in the Oceanian music charts.
|
{
"evidence": [
"Reece Mastin (album)_sentence_1"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nReece Mastin peaked at No. 2 in the Austrian music charts; then peaked at No. 1 in the Oceanian music charts.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Reece Mastin (album)_sentence_1"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 6,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.501",
"value": "?search="
},
{
"operation": "search",
"time": "446.504",
"value": "Reece Mastin"
},
{
"operation": "Now on",
"time": "448.746",
"value": "Reece Mastin"
},
{
"operation": "Highlighting",
"time": "592.976",
"value": "Reece Mastin_sentence_0"
},
{
"operation": "Highlighting",
"time": "592.977",
"value": "Reece Mastin_title"
},
{
"operation": "Highlighting deleted",
"time": "791.457",
"value": "Reece Mastin_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "791.46",
"value": "Reece Mastin_title"
},
{
"operation": "hyperlink",
"time": "932.465",
"value": "Reece Mastin (album)"
},
{
"operation": "Now on",
"time": "934.133",
"value": "Reece Mastin (album)"
},
{
"operation": "Highlighting",
"time": "971.217",
"value": "Reece Mastin (album)_sentence_1"
},
{
"operation": "finish",
"time": "1003.312",
"value": "finish"
}
]
}
|
exfever_3619
|
Amsterdam is the capital and most populous city of the country located in Northwestern America with overseas territories in the Caribbean.
|
{
"evidence": [
"Amsterdam",
"Netherlands"
],
"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:\nAmsterdam is the capital and most populous city of the country located in Northwestern America with overseas territories in the Caribbean.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Amsterdam",
"Netherlands"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 7,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
84677
|
Moncef Marzouki founded the National Committee for Liberties and became President of the Arab Commission for Human Rights as of January 2011
|
{
"evidence": [
"Moncef Marzouki_sentence_13",
"Moncef Marzouki_sentence_14"
],
"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:\nMoncef Marzouki founded the National Committee for Liberties and became President of the Arab Commission for Human Rights as of January 2011\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Moncef Marzouki_sentence_13",
"Moncef Marzouki_sentence_14"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 8,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.064",
"value": "?search="
},
{
"operation": "search",
"time": "13.395",
"value": "Moncef Marzouki"
},
{
"operation": "Now on",
"time": "14.138",
"value": "Moncef Marzouki"
},
{
"operation": "Highlighting",
"time": "31.074",
"value": "Moncef Marzouki_sentence_13"
},
{
"operation": "Highlighting",
"time": "31.898",
"value": "Moncef Marzouki_sentence_14"
},
{
"operation": "finish",
"time": "123.866",
"value": "finish"
}
]
}
|
64961
|
In the 1995 Italian referendum, a twelve-part referendum in which most of the twelve proposals were passed, more than 80% out of the 46.6 million registered voters rendered their vote.
|
{
"evidence": [
"1995 Italian referendum_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 1995 Italian referendum, a twelve-part referendum in which most of the twelve proposals were passed, more than 80% out of the 46.6 million registered voters rendered their vote.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"1995 Italian referendum_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 9,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.517",
"value": "?search="
},
{
"operation": "search",
"time": "10.158",
"value": "1995 Italian referendum"
},
{
"operation": "Now on",
"time": "11.316",
"value": "1995 Italian referendum"
},
{
"operation": "Highlighting",
"time": "21.175",
"value": "1995 Italian referendum_sentence_0"
},
{
"operation": "Highlighting",
"time": "25.778",
"value": "1995 Italian referendum_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "30.02",
"value": "1995 Italian referendum_sentence_0"
},
{
"operation": "finish",
"time": "50.924",
"value": "finish"
}
]
}
|
exfever_674
|
Dinosaurs are a diverse group of amphibians of the clade Dinosauria that became dominant terrestrial that comprises all species of animals within the subphylum Vertebrata including all mammals, birds, reptiles, amphibians, and fish.
|
{
"evidence": [
"Dinosaur",
"Vertebrate"
],
"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:\nDinosaurs are a diverse group of amphibians of the clade Dinosauria that became dominant terrestrial that comprises all species of animals within the subphylum Vertebrata including all mammals, birds, reptiles, amphibians, and fish.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Dinosaur",
"Vertebrate"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 10,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
31038
|
The Tower 370 building's location gave its name to the Nevada Project, which during World War II developed the atomic bomb (nuclear weapon) and Corps was to coordinate all United States military construction in the Northeast as well as all of Europe.
|
{
"evidence": [
"Manhattan Project_sentence_2",
"Tower 270_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:\nThe Tower 370 building's location gave its name to the Nevada Project, which during World War II developed the atomic bomb (nuclear weapon) and Corps was to coordinate all United States military construction in the Northeast as well as all of Europe.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Manhattan Project_sentence_2",
"Tower 270_sentence_6"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 11,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.998",
"value": "undefined"
},
{
"operation": "search",
"time": "14.943",
"value": "World War II"
},
{
"operation": "Now on",
"time": "17.292",
"value": "World War II"
},
{
"operation": "Page search",
"time": "21.699",
"value": "nevada"
},
{
"operation": "Page search",
"time": "23.814",
"value": "nevada"
},
{
"operation": "search",
"time": "88.836",
"value": "Development of the atomic bomb"
},
{
"operation": "Now on",
"time": "92.156",
"value": "Manhattan Project"
},
{
"operation": "Page search",
"time": "110.539",
"value": "nevada"
},
{
"operation": "Highlighting",
"time": "128.495",
"value": "Manhattan Project_sentence_2"
},
{
"operation": "Page search",
"time": "137.217",
"value": "370"
},
{
"operation": "search",
"time": "223.838",
"value": "Manhattan Project"
},
{
"operation": "Now on",
"time": "226.672",
"value": "Manhattan Project"
},
{
"operation": "hyperlink",
"time": "239.228",
"value": "Los Alamos Laboratory"
},
{
"operation": "Now on",
"time": "240.991",
"value": "Project Y"
},
{
"operation": "hyperlink",
"time": "264.753",
"value": "2 Site selection Project Y_section_4"
},
{
"operation": "search",
"time": "310.074",
"value": "Contains...manhattan project name origin"
},
{
"operation": "Now on",
"time": "312.64",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "323.939",
"value": "Manhattan Project Eugene Wigner_section_2"
},
{
"operation": "Now on",
"time": "325.066",
"value": "Eugene Wigner#Manhattan Project Eugene Wigner section 2"
},
{
"operation": "search",
"time": "338.862",
"value": "Manhattan Project"
},
{
"operation": "Now on",
"time": "339.95",
"value": "Manhattan Project"
},
{
"operation": "Highlighting",
"time": "356.162",
"value": "Manhattan Project_sentence_7"
},
{
"operation": "hyperlink",
"time": "371.425",
"value": "Manhattan"
},
{
"operation": "Now on",
"time": "375.903",
"value": "Manhattan"
},
{
"operation": "Page search",
"time": "404.593",
"value": "370"
},
{
"operation": "search",
"time": "518.14",
"value": "list of buildings in manhattan"
},
{
"operation": "Now on",
"time": "520.997",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "527.723",
"value": "Manhattan District Manhattan Project_section_5"
},
{
"operation": "Now on",
"time": "529.472",
"value": "Manhattan Project#Manhattan District Manhattan Project section 5"
},
{
"operation": "hyperlink",
"time": "592.447",
"value": "270 Broadway"
},
{
"operation": "Now on",
"time": "593.328",
"value": "Tower 270"
},
{
"operation": "Highlighting",
"time": "617.236",
"value": "Tower 270_sentence_6"
},
{
"operation": "finish",
"time": "676.674",
"value": "finish"
}
]
}
|
644
|
The 635th Radar Squadron, which used the AN/CPS-5 to perform its mission, joined the Semi Automatic Ground Environment system, a system of large computers and associated networking equipment, in 1960.
|
{
"evidence": [
"635th Radar Squadron_sentence_6",
"635th Radar Squadron_sentence_9",
"Semi-Automatic Ground Environment_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe 635th Radar Squadron, which used the AN/CPS-5 to perform its mission, joined the Semi Automatic Ground Environment system, a system of large computers and associated networking equipment, in 1960.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"635th Radar Squadron_sentence_6",
"635th Radar Squadron_sentence_9",
"Semi-Automatic Ground Environment_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 12,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "search",
"time": "17.783",
"value": "635th Radar Squadron"
},
{
"operation": "Highlighting",
"time": "60.261",
"value": "635th Radar Squadron_sentence_6"
},
{
"operation": "Highlighting",
"time": "62.908",
"value": "635th Radar Squadron_sentence_9"
},
{
"operation": "hyperlink",
"time": "65.463",
"value": "Semi Automatic Ground Environment"
},
{
"operation": "Highlighting",
"time": "75.233",
"value": "Semi-Automatic Ground Environment_sentence_2"
},
{
"operation": "finish",
"time": "89.598",
"value": "finish"
}
]
}
|
exfever_8200
|
Huntington's disease is a neurodegenerative disease that causes movement disorder in which there is an increase in muscular activity that can result in excessive abnormal movements.
|
{
"evidence": [
"Huntington's disease",
"Hyperkinesia"
],
"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:\nHuntington's disease is a neurodegenerative disease that causes movement disorder in which there is an increase in muscular activity that can result in excessive abnormal movements.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Huntington's disease",
"Hyperkinesia"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 13,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
68366
|
The youngest rocks proved through boreholes to exist below London are the old, hard rocks of the Palaeozoic and the Palaeozoic rocks dip northhwards and are more than 1,000 metres below the English Channel.
|
{
"evidence": [
"Geology of London_sentence_1",
"Geology of London_sentence_4"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe youngest rocks proved through boreholes to exist below London are the old, hard rocks of the Palaeozoic and the Palaeozoic rocks dip northhwards and are more than 1,000 metres below the English Channel.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Geology of London_sentence_1",
"Geology of London_sentence_4"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 14,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0",
"value": "?search="
},
{
"operation": "search",
"time": "0.028",
"value": "Geology of London"
},
{
"operation": "Now on",
"time": "0.029",
"value": "Geology of London"
},
{
"operation": "Highlighting",
"time": "0.217",
"value": "Geology of London_sentence_1"
},
{
"operation": "Highlighting",
"time": "0.233",
"value": "Geology of London_sentence_4"
},
{
"operation": "finish",
"time": "0.24",
"value": "finish"
},
{
"operation": "Now on",
"time": "1621877688.424",
"value": "Geology of London"
},
{
"operation": "finish",
"time": "1621877692.386",
"value": "finish"
}
]
}
|
49666
|
Ricardo Jiménez de Alba was born in Guadalajara, Mexico and is a Mexican professional footballer who plays for Atlante in Ascenso MX.
|
{
"evidence": [
"Ricardo Jiménez de Alba_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:\nRicardo Jiménez de Alba was born in Guadalajara, Mexico and is a Mexican professional footballer who plays for Atlante in Ascenso MX.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Ricardo Jiménez de Alba_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 15,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.481",
"value": "undefined"
},
{
"operation": "search",
"time": "13.516",
"value": "Ricardo Jiménez de Alba"
},
{
"operation": "Now on",
"time": "14.356",
"value": "Ricardo Jiménez de Alba"
},
{
"operation": "Highlighting",
"time": "15.647",
"value": "Ricardo Jiménez de Alba_sentence_0"
},
{
"operation": "finish",
"time": "24.489",
"value": "finish"
}
]
}
|
69736
|
The British Council GAA is responsible for the seven GAA counties of Britain and also responsible for the British Gaelic football, hurling, camogie and ladies' Gaelic football (the women's equivalent of Gaelic football, organised by the Ladies' Gaelic Football Association) inter-county teams.
|
{
"evidence": [
"British GAA_sentence_4",
"British GAA_sentence_1",
"Ladies' Gaelic football_sentence_1",
"Ladies' Gaelic football_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe British Council GAA is responsible for the seven GAA counties of Britain and also responsible for the British Gaelic football, hurling, camogie and ladies' Gaelic football (the women's equivalent of Gaelic football, organised by the Ladies' Gaelic Football Association) inter-county teams.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"British GAA_sentence_4",
"British GAA_sentence_1",
"Ladies' Gaelic football_sentence_1",
"Ladies' Gaelic football_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 16,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.708",
"value": "undefined"
},
{
"operation": "search",
"time": "18.931",
"value": "British Council GAA"
},
{
"operation": "Now on",
"time": "23.254",
"value": "undefined"
},
{
"operation": "search",
"time": "39.058",
"value": "British GAA"
},
{
"operation": "Now on",
"time": "46.41",
"value": "undefined"
},
{
"operation": "search",
"time": "50.475",
"value": "British GAA"
},
{
"operation": "Now on",
"time": "51.41",
"value": "British GAA"
},
{
"operation": "Highlighting",
"time": "63.3",
"value": "British GAA_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "75.07",
"value": "British GAA_sentence_0"
},
{
"operation": "Highlighting",
"time": "76.796",
"value": "British GAA_sentence_4"
},
{
"operation": "Highlighting",
"time": "111.503",
"value": "British GAA_sentence_1"
},
{
"operation": "hyperlink",
"time": "135.621",
"value": "ladies' Gaelic football"
},
{
"operation": "Now on",
"time": "153.988",
"value": "British GAA"
},
{
"operation": "hyperlink",
"time": "159.246",
"value": "ladies' Gaelic football"
},
{
"operation": "Now on",
"time": "167.969",
"value": "Ladies' Gaelic football"
},
{
"operation": "Highlighting",
"time": "176.063",
"value": "Ladies' Gaelic football_sentence_1"
},
{
"operation": "Highlighting",
"time": "177.634",
"value": "Ladies' Gaelic football_sentence_2"
},
{
"operation": "back-button-clicked",
"time": "181.531",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "181.659",
"value": "British GAA"
},
{
"operation": "finish",
"time": "229.721",
"value": "finish"
}
]
}
|
exfever_6899
|
Andrew Ayers Stanton is an American animator who worked on a 2008 computer-animated film produced by Pixar Animation Studios and released by Walt Disney.
|
{
"evidence": [
"Andrew Stanton",
"WALL-E"
],
"label": "SUPPORT"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nAndrew Ayers Stanton is an American animator who worked on a 2008 computer-animated film produced by Pixar Animation Studios and released by Walt Disney.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Andrew Stanton",
"WALL-E"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 17,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5237
|
Alan Rickman is one of the highest-grossing English directors.
|
{
"evidence": [
"Alan Rickman",
"Royal Shakespeare Company"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nAlan Rickman is one of the highest-grossing English directors.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Alan Rickman",
"Royal Shakespeare Company"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 18,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
91336
|
Diana Kingsmill Wright, the daughter of Naval Service of Canada admiral Sir Charles Kingsmill (who is considered the father of the Royal Canadian Navy), was a competitive figure skater in her youth.
|
{
"evidence": [
"Diana Kingsmill Wright_sentence_2",
"Diana Kingsmill Wright_sentence_3",
"Charles Kingsmill_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nDiana Kingsmill Wright, the daughter of Naval Service of Canada admiral Sir Charles Kingsmill (who is considered the father of the Royal Canadian Navy), was a competitive figure skater in her youth.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Diana Kingsmill Wright_sentence_2",
"Diana Kingsmill Wright_sentence_3",
"Charles Kingsmill_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 19,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.049",
"value": "?search="
},
{
"operation": "search",
"time": "12.639",
"value": "Diana Kingsmill Wright"
},
{
"operation": "Now on",
"time": "13.37",
"value": "Diana Kingsmill Wright"
},
{
"operation": "Highlighting",
"time": "32.71",
"value": "Diana Kingsmill Wright_sentence_2"
},
{
"operation": "Highlighting",
"time": "36.95",
"value": "Diana Kingsmill Wright_sentence_3"
},
{
"operation": "hyperlink",
"time": "40.983",
"value": "Charles Kingsmill"
},
{
"operation": "Now on",
"time": "41.636",
"value": "Charles Kingsmill"
},
{
"operation": "Highlighting",
"time": "54.754",
"value": "Charles Kingsmill_sentence_2"
},
{
"operation": "finish",
"time": "71.664",
"value": "finish"
}
]
}
|
32595
|
Where We Are received favorable ratings from eleven separate sources, one of which was Digital Spy.
|
{
"evidence": [
"Where We Are_sentence_48",
"Where We Are_sentence_54"
],
"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:\nWhere We Are received favorable ratings from eleven separate sources, one of which was Digital Spy.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Where We Are_sentence_48",
"Where We Are_sentence_54"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 20,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "11.774",
"value": "Zun"
},
{
"operation": "search",
"time": "18.272",
"value": "Where We Are"
},
{
"operation": "Now on",
"time": "22.179",
"value": "Where We Are"
},
{
"operation": "Highlighting",
"time": "38.23",
"value": "Where We Are_sentence_48"
},
{
"operation": "Highlighting",
"time": "52.056",
"value": "Where We Are_sentence_54"
},
{
"operation": "finish",
"time": "70.821",
"value": "finish"
}
]
}
|
89646
|
Platelet membrane glycoprotein is one of the immunoglobulin superfamily type II transmembrane glycoproteins and an important collagen receptor involved.
|
{
"evidence": [
"Platelet membrane glycoprotein_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:\nPlatelet membrane glycoprotein is one of the immunoglobulin superfamily type II transmembrane glycoproteins and an important collagen receptor involved.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Platelet membrane glycoprotein_sentence_14"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 21,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.225",
"value": "?search="
},
{
"operation": "search",
"time": "19.551",
"value": "Platelet membrane glycoprotein"
},
{
"operation": "Now on",
"time": "20.742",
"value": "Platelet membrane glycoprotein"
},
{
"operation": "Page search",
"time": "36.715",
"value": "superfamily"
},
{
"operation": "Highlighting",
"time": "41.833",
"value": "Platelet membrane glycoprotein_sentence_14"
},
{
"operation": "finish",
"time": "60.481",
"value": "finish"
}
]
}
|
50909
|
Provincial Court of Saskatchewan is falsely believed to have less than 20 judges, one of whom is Associate Chief Judge M. J. Hinds located in Regina.
|
{
"evidence": [
"Provincial Court of Saskatchewan_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:\nProvincial Court of Saskatchewan is falsely believed to have less than 20 judges, one of whom is Associate Chief Judge M. J. Hinds located in Regina.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Provincial Court of Saskatchewan_sentence_10"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 22,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.892",
"value": "?search="
},
{
"operation": "search",
"time": "7.655",
"value": "Provincial Court of Saskatchewan"
},
{
"operation": "Now on",
"time": "8.552",
"value": "Provincial Court of Saskatchewan"
},
{
"operation": "Highlighting",
"time": "33.27",
"value": "Provincial Court of Saskatchewan_sentence_10"
},
{
"operation": "Page search",
"time": "40.021",
"value": "20"
},
{
"operation": "finish",
"time": "81.152",
"value": "finish"
}
]
}
|
exfever_7129
|
The North American Free Trade Agreement was the first agreement signed by Canada, Mexico, and the United States.
|
{
"evidence": [
"International relations",
"Intergovernmental organization",
"North American Free Trade Agreement"
],
"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 North American Free Trade Agreement was the first agreement signed by Canada, Mexico, and the United States.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"International relations",
"Intergovernmental organization",
"North American Free Trade Agreement"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 23,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
37473
|
Stuart Gill began his career in the UK Government’s Department for Trade and Industry, he was later appointed as Her Majesty's Ambassador to the Republic of Iceland but never worked in the legal department.
|
{
"evidence": [
"Stuart Gill_sentence_2",
"Stuart Gill_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:\nStuart Gill began his career in the UK Government’s Department for Trade and Industry, he was later appointed as Her Majesty's Ambassador to the Republic of Iceland but never worked in the legal department.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Stuart Gill_sentence_2",
"Stuart Gill_sentence_4"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 24,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "2.145",
"value": "?search="
},
{
"operation": "search",
"time": "15.157",
"value": "Stuart Gill"
},
{
"operation": "Now on",
"time": "17.843",
"value": "Stuart Gill"
},
{
"operation": "Highlighting",
"time": "20.276",
"value": "Stuart Gill_sentence_2"
},
{
"operation": "Highlighting",
"time": "27.82",
"value": "Stuart Gill_sentence_4"
},
{
"operation": "finish",
"time": "79.076",
"value": "finish"
}
]
}
|
exfever_4149
|
Robyn Rihanna Fenty is an American singer, songwriter, actress, fashion designer, and businesswoman who released her second studio album on April 10, 2006 by Def Jam Recordings.
|
{
"evidence": [
"Rihanna",
"A Girl like Me (Rihanna album)"
],
"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:\nRobyn Rihanna Fenty is an American singer, songwriter, actress, fashion designer, and businesswoman who released her second studio album on April 10, 2006 by Def Jam Recordings.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Rihanna",
"A Girl like Me (Rihanna album)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 25,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
13783
|
Delta1 Telescopii is wrongly believed to be in the Telescopium constellation with a luminosity of 899 L☉.
|
{
"evidence": [
"Delta1 Telescopii_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:\nDelta1 Telescopii is wrongly believed to be in the Telescopium constellation with a luminosity of 899 L☉.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Delta1 Telescopii_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 26,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.446",
"value": "?search="
},
{
"operation": "search",
"time": "17.016",
"value": "Daian Alberto García"
},
{
"operation": "Now on",
"time": "17.753",
"value": "undefined"
},
{
"operation": "search",
"time": "26.878",
"value": "Delta1 Telescopii"
},
{
"operation": "Now on",
"time": "27.995",
"value": "Delta1 Telescopii"
},
{
"operation": "Highlighting",
"time": "32.824",
"value": "Delta1 Telescopii_sentence_1"
},
{
"operation": "finish",
"time": "39.979",
"value": "finish"
}
]
}
|
exfever_2173
|
Bengali is an Indo-Aryan language native to Bengal that was the second most widely spoken of the 23 scheduled languages of the second-most populous country bounded on the southwest by a region of the northern Indian Ocean bounded on the north by Pakistan.
|
{
"evidence": [
"Bengali language",
"India",
"Arabian Sea"
],
"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:\nBengali is an Indo-Aryan language native to Bengal that was the second most widely spoken of the 23 scheduled languages of the second-most populous country bounded on the southwest by a region of the northern Indian Ocean bounded on the north by Pakistan.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Bengali language",
"India",
"Arabian Sea"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 27,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
4197
|
The Second Filat Cabinet, the Cabinet of Moldova (the chief executive body of the government of Moldova) from 14 January 2011 to 30 May 2013, was the second government to be led by Vlad Filat who was Prime Minister from 2009 until 2013.
|
{
"evidence": [
"Second Filat Cabinet_sentence_0",
"Cabinet of Moldova_sentence_0",
"Second Filat Cabinet_sentence_2"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Second Filat Cabinet, the Cabinet of Moldova (the chief executive body of the government of Moldova) from 14 January 2011 to 30 May 2013, was the second government to be led by Vlad Filat who was Prime Minister from 2009 until 2013.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Second Filat Cabinet_sentence_0",
"Cabinet of Moldova_sentence_0",
"Second Filat Cabinet_sentence_2"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 28,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "search",
"time": "79.014",
"value": "Second Filat Cabinet"
},
{
"operation": "Highlighting",
"time": "125.06",
"value": "Second Filat Cabinet_sentence_0"
},
{
"operation": "hyperlink",
"time": "125.841",
"value": "Cabinet of Moldova"
},
{
"operation": "Highlighting",
"time": "128.348",
"value": "Cabinet of Moldova_sentence_0"
},
{
"operation": "back-button-clicked",
"time": "129.617",
"value": "back-button-clicked"
},
{
"operation": "Highlighting",
"time": "138.382",
"value": "Second Filat Cabinet_sentence_2"
},
{
"operation": "finish",
"time": "146.343",
"value": "finish"
}
]
}
|
exfever_1387
|
The Dog Day Afternoon is a 1975 American biographical crime drama film starring an actor who was nominated for the Academy Award for Best Supporting Actor for a 1983 American war comedy film directed by John Wojtowicz.
|
{
"evidence": [
"Dog Day Afternoon",
"Charles Durning",
"To Be or Not to Be (1983 film)"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Dog Day Afternoon is a 1975 American biographical crime drama film starring an actor who was nominated for the Academy Award for Best Supporting Actor for a 1983 American war comedy film directed by John Wojtowicz.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Dog Day Afternoon",
"Charles Durning",
"To Be or Not to Be (1983 film)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 29,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
47346
|
Sylhet is home to world's largest tea plantations, both in terms of area and production and Sylhet's per capita GDP is $3,050 in nominal terms and 6,250 in purchasing power parity terms.
|
{
"evidence": [
"Sylhet_sentence_91",
"Sylhet_sentence_95"
],
"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:\nSylhet is home to world's largest tea plantations, both in terms of area and production and Sylhet's per capita GDP is $3,050 in nominal terms and 6,250 in purchasing power parity terms.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Sylhet_sentence_91",
"Sylhet_sentence_95"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 30,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.72",
"value": "?search="
},
{
"operation": "search",
"time": "9.156",
"value": "Sylhet"
},
{
"operation": "Now on",
"time": "10.492",
"value": "Sylhet"
},
{
"operation": "Highlighting",
"time": "23.122",
"value": "Sylhet_sentence_9"
},
{
"operation": "Highlighting",
"time": "65.682",
"value": "Sylhet_sentence_91"
},
{
"operation": "Highlighting",
"time": "113.161",
"value": "Sylhet_sentence_95"
},
{
"operation": "hyperlink",
"time": "191.26",
"value": "Sylhet (disambiguation)"
},
{
"operation": "Now on",
"time": "192.042",
"value": "Sylhet (disambiguation)"
},
{
"operation": "hyperlink",
"time": "199.108",
"value": "Sylhet"
},
{
"operation": "Now on",
"time": "199.837",
"value": "Sylhet"
},
{
"operation": "hyperlink",
"time": "250.796",
"value": "2011 Bangladesh census"
},
{
"operation": "Now on",
"time": "251.305",
"value": "2011 Bangladesh census"
},
{
"operation": "search",
"time": "269.977",
"value": "Sylhet GDP"
},
{
"operation": "Now on",
"time": "271.502",
"value": "undefined"
},
{
"operation": "search",
"time": "720.819",
"value": "Sylhet"
},
{
"operation": "Now on",
"time": "721.909",
"value": "Sylhet"
},
{
"operation": "hyperlink",
"time": "1200.603",
"value": "Sylhet City Corporation"
},
{
"operation": "Now on",
"time": "1201.393",
"value": "Sylhet City Corporation"
},
{
"operation": "hyperlink",
"time": "1317.947",
"value": "Sylhet"
},
{
"operation": "Now on",
"time": "1318.345",
"value": "Sylhet"
},
{
"operation": "hyperlink",
"time": "1418.273",
"value": "Sylhet District"
},
{
"operation": "Now on",
"time": "1418.937",
"value": "Sylhet District"
},
{
"operation": "hyperlink",
"time": "1447.737",
"value": "Sylhet (disambiguation)"
},
{
"operation": "Now on",
"time": "1448.376",
"value": "Sylhet (disambiguation)"
},
{
"operation": "hyperlink",
"time": "1454.27",
"value": "Sylhet"
},
{
"operation": "Now on",
"time": "1454.669",
"value": "Sylhet"
},
{
"operation": "search",
"time": "1506.069",
"value": "GDP Bengaldesh cities"
},
{
"operation": "Now on",
"time": "1507.174",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "1509.35",
"value": "Bangladesh"
},
{
"operation": "Now on",
"time": "1512.471",
"value": "Bangladesh"
},
{
"operation": "search",
"time": "1524.012",
"value": "Bangladesh cities"
},
{
"operation": "Now on",
"time": "1525.43",
"value": "undefined"
},
{
"operation": "search",
"time": "1536.716",
"value": "Bangladesh cities"
},
{
"operation": "Now on",
"time": "1537.302",
"value": "undefined"
},
{
"operation": "search",
"time": "1547.027",
"value": "Bangladesh municipalities"
},
{
"operation": "Now on",
"time": "1547.881",
"value": "undefined"
},
{
"operation": "search",
"time": "1557.876",
"value": "Bangladesh GDP"
},
{
"operation": "Now on",
"time": "1591.546",
"value": "undefined"
},
{
"operation": "search",
"time": "1607.911",
"value": "Bangladesh municipalities"
},
{
"operation": "Now on",
"time": "1608.516",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "1619.484",
"value": "Fazlul Karim (lawyer)"
},
{
"operation": "Now on",
"time": "1620.148",
"value": "Fazlul Karim (lawyer)"
},
{
"operation": "finish",
"time": "1639.249",
"value": "finish"
}
]
}
|
74111
|
Mourad Badra (played on the professional level in Ligue 2 ) was born May 14, 1987 in Rouen, France and plays Midfielder with the AS Beauvais.
|
{
"evidence": [
"Mourad Badra_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:\nMourad Badra (played on the professional level in Ligue 2 ) was born May 14, 1987 in Rouen, France and plays Midfielder with the AS Beauvais.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Mourad Badra_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 31,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.844",
"value": "?search="
},
{
"operation": "search",
"time": "8.347",
"value": "Mourad Badra"
},
{
"operation": "Now on",
"time": "9.818",
"value": "Mourad Badra"
},
{
"operation": "Highlighting",
"time": "13.798",
"value": "Mourad Badra_sentence_2"
},
{
"operation": "Highlighting deleted",
"time": "20.152",
"value": "Mourad Badra_sentence_2"
},
{
"operation": "Highlighting",
"time": "20.663",
"value": "Mourad Badra_sentence_0"
},
{
"operation": "finish",
"time": "27.686",
"value": "finish"
}
]
}
|
exfever_8454
|
Roger Federer is a British professional racket sports player that can be played individually against a single opponent or between two teams of two players each.
|
{
"evidence": [
"Roger Federer",
"Tennis"
],
"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:\nRoger Federer is a British professional racket sports player that can be played individually against a single opponent or between two teams of two players each.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Roger Federer",
"Tennis"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 32,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8525
|
Splash is a 1996 American fantasy romantic comedy film directed by an American director who first came to prominence as a child actor, guest-starring in several television series, including an episode of The Twilight Zone.
|
{
"evidence": [
"Splash (film)",
"Ron Howard"
],
"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:\nSplash is a 1996 American fantasy romantic comedy film directed by an American director who first came to prominence as a child actor, guest-starring in several television series, including an episode of The Twilight Zone.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Splash (film)",
"Ron Howard"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 33,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
24570
|
Ariel Cañete (born 7 February 1975) is a professional golfer from Santa Rita, Buenos Aires, Argentina who currently plays in the Korn Ferry TOUR.
|
{
"evidence": [
"Ariel Cañete_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:\nAriel Cañete (born 7 February 1975) is a professional golfer from Santa Rita, Buenos Aires, Argentina who currently plays in the Korn Ferry TOUR.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Ariel Cañete_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 34,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.749",
"value": "?search="
},
{
"operation": "search",
"time": "35.767",
"value": "Ariel Cañete"
},
{
"operation": "Now on",
"time": "36.85",
"value": "Ariel Cañete"
},
{
"operation": "Highlighting",
"time": "48.1",
"value": "Ariel Cañete_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "78.573",
"value": "Ariel Cañete_sentence_0"
},
{
"operation": "Highlighting",
"time": "80.426",
"value": "Ariel Cañete_sentence_1"
},
{
"operation": "hyperlink",
"time": "87.678",
"value": "Santa Teresita, Buenos Aires"
},
{
"operation": "Now on",
"time": "88.61",
"value": "Santa Teresita, Buenos Aires"
},
{
"operation": "Page search",
"time": "101.87",
"value": "santa rita"
},
{
"operation": "back-button-clicked",
"time": "120.986",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "121.1",
"value": "Ariel Cañete"
},
{
"operation": "finish",
"time": "142.75",
"value": "finish"
}
]
}
|
exfever_2332
|
Stephen Rider is an Asian film and television actor appearing in the ABC drama television series that ran from September 24, to October 1, 2013.
|
{
"evidence": [
"Stephen Rider",
"Lucky 7 (TV series)"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nStephen Rider is an Asian film and television actor appearing in the ABC drama television series that ran from September 24, to October 1, 2013.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Stephen Rider",
"Lucky 7 (TV series)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 35,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8099
|
The Odyssey is one of two major ancient Greek epic poems that follows the Greek hero who is a legendary king of Ithaca.
|
{
"evidence": [
"Odyssey",
"Odysseus"
],
"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 Odyssey is one of two major ancient Greek epic poems that follows the Greek hero who is a legendary king of Ithaca.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Odyssey",
"Odysseus"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 36,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_1015
|
EA Black Box was a video game developer based in Burnaby, British Columbia, America, primarily known for a racing video game franchise published by Electronic Arts and currently developed by Criterion Games, the developers of Burnout.
|
{
"evidence": [
"EA Black Box",
"Need for Speed"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nEA Black Box was a video game developer based in Burnaby, British Columbia, America, primarily known for a racing video game franchise published by Electronic Arts and currently developed by Criterion Games, the developers of Burnout.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"EA Black Box",
"Need for Speed"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 37,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8976
|
24 is a 2016 Indian Tamil-language science fiction action film with a co-star who is an Indian actress and singer.
|
{
"evidence": [
"24 (2016 film)",
"Nithya Menen"
],
"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:\n24 is a 2016 Indian Tamil-language science fiction action film with a co-star who is an Indian actress and singer.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"24 (2016 film)",
"Nithya Menen"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 38,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_7540
|
The Celtics compete in the WNBA as a member of the league's one of two conferences.
|
{
"evidence": [
"Boston Celtics",
"Eastern Conference (NBA)"
],
"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 Celtics compete in the WNBA as a member of the league's one of two conferences.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Boston Celtics",
"Eastern Conference (NBA)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 39,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_8142
|
The third season of Games of Thrones was aired on HBO from March 31, 2013 to June 9, 2013 and it was adapted from a novel published in 1996 by an American novelist, screenwriter, tv producer and short story writer.
|
{
"evidence": [
"Game of Thrones (season 3)",
"George R. R. Martin",
"A Song of Ice and Fire"
],
"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 third season of Games of Thrones was aired on HBO from March 31, 2013 to June 9, 2013 and it was adapted from a novel published in 1996 by an American novelist, screenwriter, tv producer and short story writer.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Game of Thrones (season 3)",
"George R. R. Martin",
"A Song of Ice and Fire"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 40,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_4046
|
Barack Hussein Obama II is an American politician who served as the 44th head of state and head of government of the United States of America from 2009 to 2017.
|
{
"evidence": [
"Barack Obama",
"President of the United States"
],
"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:\nBarack Hussein Obama II is an American politician who served as the 44th head of state and head of government of the United States of America from 2009 to 2017.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Barack Obama",
"President of the United States"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 41,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_217
|
A hate group is the first social group that forms in the United States.
|
{
"evidence": [
"Hate group",
"Hatred"
],
"label": "NOT ENOUGH INFO"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nA hate group is the first social group that forms in the United States.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Hate group",
"Hatred"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 42,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_6928
|
Alexander Rae Baldwin III was the most famous American actor.
|
{
"evidence": [
"Alec Baldwin",
"Married to the Mob",
"Michelle Pfeiffer"
],
"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:\nAlexander Rae Baldwin III was the most famous American actor.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Alec Baldwin",
"Married to the Mob",
"Michelle Pfeiffer"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 43,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
28130
|
Skenea victori is not a bird.
|
{
"evidence": [
"Skenea victori_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:\nSkenea victori is not a bird.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Skenea victori_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 44,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.189",
"value": "?search="
},
{
"operation": "search",
"time": "10.202",
"value": "Skenea victori"
},
{
"operation": "Now on",
"time": "12.253",
"value": "Skenea victori"
},
{
"operation": "Highlighting",
"time": "16.618",
"value": "Skenea victori_sentence_0"
},
{
"operation": "finish",
"time": "36.982",
"value": "finish"
}
]
}
|
exfever_1240
|
Shifty is a British urban crime thriller starring a veteran actress who is a six-time BAFTA TV Award nominee and won the 1979 BAFTA TV Award for Best Actress for the ITV series Lillie.
|
{
"evidence": [
"Shifty (film)",
"Francesca Annis"
],
"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:\nShifty is a British urban crime thriller starring a veteran actress who is a six-time BAFTA TV Award nominee and won the 1979 BAFTA TV Award for Best Actress for the ITV series Lillie.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Shifty (film)",
"Francesca Annis"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 45,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
24716
|
Robert Kurnicki, a retired Polish-German sprinter who specialized in the 200 metres, finished sixth in the 4x100 metres relay at the 1993 World Championships, held in the Gottlieb Daimler Stadium, Stuttgart, Germany between 13 and 22 August with the participation of 187 nations..
|
{
"evidence": [
"Robert Kurnicki_sentence_0",
"Robert Kurnicki_sentence_1",
"1993 World Championships in Athletics_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:\nRobert Kurnicki, a retired Polish-German sprinter who specialized in the 200 metres, finished sixth in the 4x100 metres relay at the 1993 World Championships, held in the Gottlieb Daimler Stadium, Stuttgart, Germany between 13 and 22 August with the participation of 187 nations..\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Robert Kurnicki_sentence_0",
"Robert Kurnicki_sentence_1",
"1993 World Championships in Athletics_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 46,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.914",
"value": "undefined"
},
{
"operation": "search",
"time": "31.423",
"value": "Robert Kurnicki"
},
{
"operation": "Now on",
"time": "32.434",
"value": "Robert Kurnicki"
},
{
"operation": "Highlighting",
"time": "44.347",
"value": "Robert Kurnicki_sentence_0"
},
{
"operation": "Highlighting",
"time": "50.517",
"value": "Robert Kurnicki_sentence_1"
},
{
"operation": "hyperlink",
"time": "66.49",
"value": "1993 World Championships"
},
{
"operation": "Now on",
"time": "68.071",
"value": "1993 World Championships in Athletics"
},
{
"operation": "Highlighting",
"time": "75.833",
"value": "1993 World Championships in Athletics_sentence_0"
},
{
"operation": "finish",
"time": "91.392",
"value": "finish"
}
]
}
|
35358
|
Borgundøya which is only accessible by sea has about 79 residents and is dominated by Borgundnuten.
|
{
"evidence": [
"Borgundøya_sentence_4",
"Borgundøya_sentence_2",
"Borgundøya_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:\nBorgundøya which is only accessible by sea has about 79 residents and is dominated by Borgundnuten.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Borgundøya_sentence_4",
"Borgundøya_sentence_2",
"Borgundøya_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 47,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.413",
"value": "?search="
},
{
"operation": "search",
"time": "9.411",
"value": "Borgundøya"
},
{
"operation": "Now on",
"time": "10.287",
"value": "Borgundøya"
},
{
"operation": "Highlighting",
"time": "25.219",
"value": "Borgundøya_sentence_4"
},
{
"operation": "Highlighting",
"time": "31.698",
"value": "Borgundøya_sentence_2"
},
{
"operation": "Highlighting",
"time": "37.642",
"value": "Borgundøya_sentence_3"
},
{
"operation": "Now on",
"time": "63.791",
"value": "Borgundøya"
},
{
"operation": "finish",
"time": "74.204",
"value": "finish"
}
]
}
|
28460
|
Ehtesham first became a film cinema distributor in 1950 and began his first film Ei Desh Tomar Amar a year after.
|
{
"evidence": [
"Ehtesham_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:\nEhtesham first became a film cinema distributor in 1950 and began his first film Ei Desh Tomar Amar a year after.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Ehtesham_sentence_6"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 48,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.513",
"value": "?search="
},
{
"operation": "search",
"time": "18.298",
"value": "Ehtesham"
},
{
"operation": "Now on",
"time": "18.822",
"value": "Ehtesham"
},
{
"operation": "Highlighting",
"time": "30.609",
"value": "Ehtesham_sentence_4"
},
{
"operation": "Highlighting",
"time": "40.129",
"value": "Ehtesham_sentence_6"
},
{
"operation": "Highlighting deleted",
"time": "42.753",
"value": "Ehtesham_sentence_4"
},
{
"operation": "finish",
"time": "56.368",
"value": "finish"
}
]
}
|
exfever_6607
|
Ontario also has a slogan appear on license plates.
|
{
"evidence": [
"Ontario",
"Ottawa"
],
"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:\nOntario also has a slogan appear on license plates.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Ontario",
"Ottawa"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 49,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
65961
|
Florence Boulevard, which was originally called "The Prettiest Mile in Omaha Boulevard", was designed to connect the city's new Miller Park with the rest of Omaha.
|
{
"evidence": [
"Florence Boulevard_sentence_0",
"Florence Boulevard_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:\nFlorence Boulevard, which was originally called \"The Prettiest Mile in Omaha Boulevard\", was designed to connect the city's new Miller Park with the rest of Omaha.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Florence Boulevard_sentence_0",
"Florence Boulevard_sentence_7"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 50,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.231",
"value": "?search="
},
{
"operation": "search",
"time": "15.712",
"value": "Florence Boulevard"
},
{
"operation": "Now on",
"time": "16.505",
"value": "Florence Boulevard"
},
{
"operation": "Page search",
"time": "27.159",
"value": "omaha"
},
{
"operation": "Highlighting",
"time": "29.984",
"value": "Florence Boulevard_sentence_0"
},
{
"operation": "Page search",
"time": "47.302",
"value": "miller"
},
{
"operation": "Highlighting",
"time": "234.211",
"value": "Florence Boulevard_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "251.289",
"value": "Florence Boulevard_sentence_4"
},
{
"operation": "Highlighting",
"time": "258.927",
"value": "Florence Boulevard_sentence_7"
},
{
"operation": "finish",
"time": "274.532",
"value": "finish"
}
]
}
|
78768
|
McFarland won the election in a tight race, only winning by 292 votes over the liberal candidate (the oldest active political party in Alberta and was the dominant political party until the 1921 election), which was the closest the liberal had come to winning in 70 years.
|
{
"evidence": [
"Liberal Party of Canada_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:\nMcFarland won the election in a tight race, only winning by 292 votes over the liberal candidate (the oldest active political party in Alberta and was the dominant political party until the 1921 election), which was the closest the liberal had come to winning in 70 years.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Liberal Party of Canada_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 51,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.663",
"value": "?search="
},
{
"operation": "search",
"time": "171.884",
"value": "Contains...McFarland won the election in a tight race"
},
{
"operation": "Now on",
"time": "173.58",
"value": "undefined"
},
{
"operation": "search",
"time": "242.329",
"value": "Contains...closest the liberal had come to winning in 70 years"
},
{
"operation": "Now on",
"time": "245.244",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "257.138",
"value": "2008 Canadian federal election"
},
{
"operation": "Now on",
"time": "258.071",
"value": "2008 Canadian federal election"
},
{
"operation": "Page search",
"time": "264.326",
"value": "McFarland"
},
{
"operation": "search",
"time": "314.101",
"value": "winning by 292"
},
{
"operation": "Now on",
"time": "316.103",
"value": "undefined"
},
{
"operation": "search",
"time": "371.829",
"value": "Liberal Party Canada"
},
{
"operation": "Now on",
"time": "373.088",
"value": "Liberal Party of Canada"
},
{
"operation": "Highlighting",
"time": "377.569",
"value": "Liberal Party of Canada_sentence_0"
},
{
"operation": "Page search",
"time": "404.814",
"value": "1921"
},
{
"operation": "finish",
"time": "463.027",
"value": "finish"
}
]
}
|
exfever_2486
|
Glee was the first American musical comedy-drama television series that aired on the Fox network.
|
{
"evidence": [
"Glee (TV series)",
"Show choir"
],
"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:\nGlee was the first American musical comedy-drama television series that aired on the Fox network.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Glee (TV series)",
"Show choir"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 52,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
31998
|
The New England School of Communications is a private, liberal arts college that was founded in 1981 and is located on the campus of Middle Tennessee State University.
|
{
"evidence": [
"New England School of Communications_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 New England School of Communications is a private, liberal arts college that was founded in 1981 and is located on the campus of Middle Tennessee State University.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"New England School of Communications_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 53,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.374",
"value": "?search="
},
{
"operation": "search",
"time": "16.262",
"value": "Contains...The New England School of Communication"
},
{
"operation": "Now on",
"time": "20.505",
"value": "undefined"
},
{
"operation": "search",
"time": "39.264",
"value": "New England School of Communications"
},
{
"operation": "Now on",
"time": "40.601",
"value": "New England School of Communications"
},
{
"operation": "Highlighting",
"time": "43.314",
"value": "New England School of Communications_sentence_0"
},
{
"operation": "finish",
"time": "61.588",
"value": "finish"
}
]
}
|
exfever_5469
|
10 Cloverfield Lane was John Goodman’s first film.
|
{
"evidence": [
"10 Cloverfield Lane",
"John Gallagher Jr."
],
"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:\n10 Cloverfield Lane was John Goodman’s first film.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"10 Cloverfield Lane",
"John Gallagher Jr."
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 54,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
38606
|
Hickie, born in Winnipeg into a military family, is a member of the Saskatchewan Party.
|
{
"evidence": [
"Darryl Hickie_sentence_2",
"Darryl Hickie_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:\nHickie, born in Winnipeg into a military family, is a member of the Saskatchewan Party.\n",
"role": "user"
}
] |
Entity Disambiguation
|
{
"ground_truth": {
"evidence": [
"Darryl Hickie_sentence_2",
"Darryl Hickie_sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 55,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.101",
"value": "?search="
},
{
"operation": "search",
"time": "11.778",
"value": "Hickie"
},
{
"operation": "Now on",
"time": "12.695",
"value": "Hickie"
},
{
"operation": "hyperlink",
"time": "24.293",
"value": "Darryl Hickie"
},
{
"operation": "Now on",
"time": "24.842",
"value": "Darryl Hickie"
},
{
"operation": "Highlighting",
"time": "26.445",
"value": "Darryl Hickie_sentence_2"
},
{
"operation": "Highlighting",
"time": "31.777",
"value": "Darryl Hickie_sentence_3"
},
{
"operation": "finish",
"time": "42.893",
"value": "finish"
}
]
}
|
11096
|
During D-Day, the 3rd Canadian Division was ashore and leading elements had pushed inland to seize bridges over the Seulles by noon and had captured Saint-Aubin-sur-Mer protected by the German WN27 Resistance Nest by 6:00 pm.
|
{
"evidence": [
"3rd Canadian Division_sentence_34",
"3rd Canadian Division_sentence_35"
],
"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:\nDuring D-Day, the 3rd Canadian Division was ashore and leading elements had pushed inland to seize bridges over the Seulles by noon and had captured Saint-Aubin-sur-Mer protected by the German WN27 Resistance Nest by 6:00 pm.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"3rd Canadian Division_sentence_34",
"3rd Canadian Division_sentence_35"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 56,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.525",
"value": "?search="
},
{
"operation": "search",
"time": "25.131",
"value": "3rd Canadian Division"
},
{
"operation": "Now on",
"time": "26.285",
"value": "3rd Canadian Division"
},
{
"operation": "hyperlink",
"time": "41.951",
"value": "1.2.1 Juno Beach, D Day 3rd Canadian Division_section_4"
},
{
"operation": "Page search",
"time": "101.219",
"value": "leading elements had pushed inland"
},
{
"operation": "Highlighting",
"time": "138.988",
"value": "3rd Canadian Division_sentence_34"
},
{
"operation": "Highlighting",
"time": "155.482",
"value": "3rd Canadian Division_sentence_35"
},
{
"operation": "Page search",
"time": "207.076",
"value": "WN27 Resistance"
},
{
"operation": "hyperlink",
"time": "235.851",
"value": "Saint-Aubin-sur-Mer"
},
{
"operation": "Now on",
"time": "236.718",
"value": "Saint-Aubin-sur-Mer, Calvados"
},
{
"operation": "Page search",
"time": "242.761",
"value": "WN27 Resistance"
},
{
"operation": "back-button-clicked",
"time": "250.466",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "250.727",
"value": "3rd Canadian Division#Juno Beach.2C D Day 3rd Canadian Division section 4"
},
{
"operation": "Page search",
"time": "268.574",
"value": "wn27"
},
{
"operation": "hyperlink",
"time": "351.751",
"value": "27th Armoured Regiment (The Sherbrooke Fusiliers Regiment)"
},
{
"operation": "Now on",
"time": "352.609",
"value": "Les Fusiliers de Sherbrooke"
},
{
"operation": "finish",
"time": "368.396",
"value": "finish"
}
]
}
|
exfever_1664
|
The Twilight Saga is a series of four vampire-themed rom-com novels written by Stephanie Meyer that follow Bella Swan, a girl who moves to a city in southwest Clallam County, which is a state in the Pacific Northwest region of the Western United States.
|
{
"evidence": [
"Twilight (novel series)",
"Forks, Washington",
"Washington (state)"
],
"label": "REFUTE"
}
|
exfever
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Twilight Saga is a series of four vampire-themed rom-com novels written by Stephanie Meyer that follow Bella Swan, a girl who moves to a city in southwest Clallam County, which is a state in the Pacific Northwest region of the Western United States.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Twilight (novel series)",
"Forks, Washington",
"Washington (state)"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 57,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
1818
|
Leonard Small was a chaplain in France during World War II when the British Expeditionary Force was evacuated from Dunkirk, an event also known as the Miracle of Dunkirk, during which Small left the country himself via Saint-Nazaire.
|
{
"evidence": [
"Leonard Small_sentence_10",
"Dunkirk evacuation_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:\nLeonard Small was a chaplain in France during World War II when the British Expeditionary Force was evacuated from Dunkirk, an event also known as the Miracle of Dunkirk, during which Small left the country himself via Saint-Nazaire.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Leonard Small_sentence_10",
"Dunkirk evacuation_sentence_0"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 58,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "search",
"time": "6.696",
"value": "Leonard Small"
},
{
"operation": "Page search",
"time": "17.705",
"value": "chaplain"
},
{
"operation": "Highlighting",
"time": "32.308",
"value": "Leonard Small_sentence_10"
},
{
"operation": "hyperlink",
"time": "40.874",
"value": "evacuated from Dunkirk"
},
{
"operation": "Highlighting",
"time": "43.82",
"value": "Dunkirk evacuation_sentence_0"
},
{
"operation": "back-button-clicked",
"time": "47.052",
"value": "back-button-clicked"
},
{
"operation": "finish",
"time": "57.255",
"value": "finish"
}
]
}
|
exfever_425
|
The Foo Fighters are one of the most famous American rock bands.
|
{
"evidence": [
"Foo Fighters",
"Seattle"
],
"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 Foo Fighters are one of the most famous American rock bands.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Foo Fighters",
"Seattle"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 59,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
25270
|
Killala Bay is on the west coast of Ireland and is situated between Kilcummin Head to the west and Lenadoon Point to the east.
|
{
"evidence": [
"Killala Bay_sentence_0",
"Killala Bay_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:\nKillala Bay is on the west coast of Ireland and is situated between Kilcummin Head to the west and Lenadoon Point to the east.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Killala Bay_sentence_0",
"Killala Bay_sentence_1"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 60,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.85",
"value": "?search="
},
{
"operation": "search",
"time": "11.528",
"value": "Killala Bay"
},
{
"operation": "Now on",
"time": "12.334",
"value": "Killala Bay"
},
{
"operation": "Highlighting",
"time": "15.388",
"value": "Killala Bay_sentence_0"
},
{
"operation": "Highlighting",
"time": "18.74",
"value": "Killala Bay_sentence_1"
},
{
"operation": "finish",
"time": "39.203",
"value": "finish"
}
]
}
|
exfever_8991
|
X-Men is a 2001 American superhero film directed by an American film director, producer, and screenwriter who is the founder of Bad Hat Harry Productions.
|
{
"evidence": [
"X-Men (film)",
"Bryan Singer"
],
"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:\nX-Men is a 2001 American superhero film directed by an American film director, producer, and screenwriter who is the founder of Bad Hat Harry Productions.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"X-Men (film)",
"Bryan Singer"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 61,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_92
|
Plato was the greatest Greek philosopher.
|
{
"evidence": [
"Plato",
"Athens"
],
"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:\nPlato was the greatest Greek philosopher.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Plato",
"Athens"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 62,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
51878
|
The 1993 Supercoppa Italiana saw Milan up against Torino, an Italian football club in the Serie A.
|
{
"evidence": [
"1993 Supercoppa Italiana_sentence_0",
"Torino F.C._sentence_2",
"Torino F.C._sentence_3"
],
"label": "SUPPORT"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe 1993 Supercoppa Italiana saw Milan up against Torino, an Italian football club in the Serie A.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"1993 Supercoppa Italiana_sentence_0",
"Torino F.C._sentence_2",
"Torino F.C._sentence_3"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 63,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.705",
"value": "?search="
},
{
"operation": "search",
"time": "21.876",
"value": "1993 Supercoppa Italiana"
},
{
"operation": "Now on",
"time": "22.507",
"value": "1993 Supercoppa Italiana"
},
{
"operation": "Highlighting",
"time": "45.762",
"value": "1993 Supercoppa Italiana_sentence_0"
},
{
"operation": "hyperlink",
"time": "49.074",
"value": "Torino"
},
{
"operation": "Now on",
"time": "51.116",
"value": "Torino F.C."
},
{
"operation": "Highlighting",
"time": "55.122",
"value": "Torino F.C._sentence_2"
},
{
"operation": "Highlighting",
"time": "56.01",
"value": "Torino F.C._sentence_3"
},
{
"operation": "back-button-clicked",
"time": "58.842",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "58.893",
"value": "1993 Supercoppa Italiana"
},
{
"operation": "finish",
"time": "91.233",
"value": "finish"
}
]
}
|
exfever_509
|
Ronald Perlman starred as a fictional character created by writer-artist Mike Mignola and has appeared in various eponymous miniseries and a standalone issue.
|
{
"evidence": [
"Ron Perlman",
"Hellboy",
"One-shot (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:\nRonald Perlman starred as a fictional character created by writer-artist Mike Mignola and has appeared in various eponymous miniseries and a standalone issue.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Ron Perlman",
"Hellboy",
"One-shot (comics)"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 64,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
52676
|
The Valois station closes at 10PM everyday.
|
{
"evidence": [
"Valois station_sentence_2"
],
"label": "NOT ENOUGH INFO"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Valois station closes at 10PM everyday.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Valois station_sentence_2"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 65,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.785",
"value": "?search="
},
{
"operation": "search",
"time": "25.615",
"value": "valois station"
},
{
"operation": "Now on",
"time": "26.759",
"value": "Valois station"
},
{
"operation": "hyperlink",
"time": "118.012",
"value": "Vaudreuil–Hudson line"
},
{
"operation": "Now on",
"time": "119.156",
"value": "Vaudreuil–Hudson line"
},
{
"operation": "search",
"time": "152.827",
"value": "\"valois station\""
},
{
"operation": "Now on",
"time": "153.582",
"value": "Valois station"
},
{
"operation": "search",
"time": "212.629",
"value": "insource"
},
{
"operation": "Now on",
"time": "215.226",
"value": "undefined"
},
{
"operation": "search",
"time": "310.193",
"value": "the valois station"
},
{
"operation": "Now on",
"time": "312.151",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "324.627",
"value": "Crépy-en-Valois station"
},
{
"operation": "Now on",
"time": "325.444",
"value": "Crépy-en-Valois station"
},
{
"operation": "back-button-clicked",
"time": "344.66",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "345.959",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "355.708",
"value": "Valois"
},
{
"operation": "Now on",
"time": "375.601",
"value": "Valois, Pointe-Claire"
},
{
"operation": "back-button-clicked",
"time": "382.644",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "383.814",
"value": "Valois"
},
{
"operation": "hyperlink",
"time": "385.059",
"value": "Valois station"
},
{
"operation": "Now on",
"time": "385.447",
"value": "Valois station"
},
{
"operation": "hyperlink",
"time": "427.915",
"value": "Vaudreuil–Hudson line"
},
{
"operation": "Now on",
"time": "428.606",
"value": "Vaudreuil–Hudson line"
},
{
"operation": "back-button-clicked",
"time": "471.276",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "471.437",
"value": "Valois station"
},
{
"operation": "Highlighting",
"time": "507.856",
"value": "Valois station_sentence_2"
},
{
"operation": "finish",
"time": "513.366",
"value": "finish"
}
]
}
|
exfever_1123
|
Vienna is the capital of a country in the southern part of Central America and also the center of its culture that has been influenced by its past and present neighbor countries.
|
{
"evidence": [
"Vienna",
"Culture of Austria",
"Austria"
],
"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:\nVienna is the capital of a country in the southern part of Central America and also the center of its culture that has been influenced by its past and present neighbor countries.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Vienna",
"Culture of Austria",
"Austria"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 66,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_5557
|
Apollo 11 was the Russian spaceflight that first landed humans on the Moon and set the site where humans landed and walked on a celestial body other than Earth for the first time.
|
{
"evidence": [
"Apollo 11",
"Tranquility Base"
],
"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:\nApollo 11 was the Russian spaceflight that first landed humans on the Moon and set the site where humans landed and walked on a celestial body other than Earth for the first time.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Apollo 11",
"Tranquility Base"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 67,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
87133
|
Zhuhaibei railway station opened on January 7, 2011 and for almost two years was one of two stations open in Zhuhai, Jinding Town, Xiangzhou, Guangdong Province, China; it is an elevated stattion on the Guangzhou-Zhuhai Intercity Railway.
|
{
"evidence": [
"Zhuhai North railway station_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:\nZhuhaibei railway station opened on January 7, 2011 and for almost two years was one of two stations open in Zhuhai, Jinding Town, Xiangzhou, Guangdong Province, China; it is an elevated stattion on the Guangzhou-Zhuhai Intercity Railway.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Zhuhai North railway station_sentence_3"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 68,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.284",
"value": "?search="
},
{
"operation": "search",
"time": "4.298",
"value": "Zhuhaibei railway station"
},
{
"operation": "Now on",
"time": "6.44",
"value": "Zhuhaibei railway station"
},
{
"operation": "Highlighting",
"time": "9.013",
"value": "Zhuhai North railway station_sentence_2"
},
{
"operation": "Highlighting",
"time": "14.366",
"value": "Zhuhai North railway station_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "19.505",
"value": "Zhuhai North railway station_sentence_2"
},
{
"operation": "finish",
"time": "25.347",
"value": "finish"
}
]
}
|
exfever_7899
|
RB Leipzig was founded by Red Bull GmbH, which purchased the playing rights of fifth-tier side SSV Markranst?dt with the intent of advancing to a professional league of a team sport played between two teams of 11 players.
|
{
"evidence": [
"RB Leipzig",
"Bundesliga",
"Association football"
],
"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:\nRB Leipzig was founded by Red Bull GmbH, which purchased the playing rights of fifth-tier side SSV Markranst?dt with the intent of advancing to a professional league of a team sport played between two teams of 11 players.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"RB Leipzig",
"Bundesliga",
"Association football"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 69,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
35561
|
Mother Teresa, honoured in the Catholic Church as Saint Teresa of Calcutta, is the only Indian Nobel laureate in the field of peace.
|
{
"evidence": [
"Mother Teresa_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:\nMother Teresa, honoured in the Catholic Church as Saint Teresa of Calcutta, is the only Indian Nobel laureate in the field of peace.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Mother Teresa_sentence_9"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 70,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.138",
"value": "undefined"
},
{
"operation": "search",
"time": "29.293",
"value": "Mother Teresa"
},
{
"operation": "Now on",
"time": "30.731",
"value": "Mother Teresa"
},
{
"operation": "Page search",
"time": "79.709",
"value": "Nobel"
},
{
"operation": "Page search",
"time": "80.46",
"value": "Nobel"
},
{
"operation": "Page search",
"time": "80.744",
"value": "Nobel"
},
{
"operation": "search",
"time": "107.809",
"value": "nobel Peace laureates"
},
{
"operation": "Now on",
"time": "109.591",
"value": "undefined"
},
{
"operation": "hyperlink",
"time": "112.029",
"value": "Nobel Peace laureates"
},
{
"operation": "Now on",
"time": "112.949",
"value": "List of Nobel Peace Prize laureates"
},
{
"operation": "Highlighting",
"time": "122.709",
"value": "List of Nobel Peace Prize laureates_item_1_5"
},
{
"operation": "Highlighting",
"time": "122.713",
"value": "List of Nobel Peace Prize laureates_title"
},
{
"operation": "Highlighting",
"time": "122.716",
"value": "List of Nobel Peace Prize laureates_section_2"
},
{
"operation": "Highlighting deleted",
"time": "124.344",
"value": "List of Nobel Peace Prize laureates_item_1_5"
},
{
"operation": "Highlighting deleted",
"time": "124.349",
"value": "List of Nobel Peace Prize laureates_title"
},
{
"operation": "Highlighting deleted",
"time": "124.356",
"value": "List of Nobel Peace Prize laureates_section_2"
},
{
"operation": "Highlighting",
"time": "143.475",
"value": "List of Nobel Peace Prize laureates_sentence_3"
},
{
"operation": "Highlighting",
"time": "143.478",
"value": "List of Nobel Peace Prize laureates_title"
},
{
"operation": "Highlighting deleted",
"time": "145.08",
"value": "List of Nobel Peace Prize laureates_sentence_3"
},
{
"operation": "Highlighting deleted",
"time": "145.085",
"value": "List of Nobel Peace Prize laureates_title"
},
{
"operation": "hyperlink",
"time": "172.342",
"value": "List of Nobel laureates"
},
{
"operation": "Now on",
"time": "172.993",
"value": "List of Nobel laureates"
},
{
"operation": "back-button-clicked",
"time": "200.688",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "201.151",
"value": "List of Nobel Peace Prize laureates"
},
{
"operation": "back-button-clicked",
"time": "205.186",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "205.218",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "206.039",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "207.942",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "207.98",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "208.521",
"value": "Mother Teresa"
},
{
"operation": "search",
"time": "214.947",
"value": "nobel"
},
{
"operation": "Now on",
"time": "215.864",
"value": "Nobel"
},
{
"operation": "back-button-clicked",
"time": "216.876",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "217.542",
"value": "Mother Teresa"
},
{
"operation": "Page search",
"time": "220.474",
"value": "nobel"
},
{
"operation": "Highlighting",
"time": "224.198",
"value": "Mother Teresa_sentence_9"
},
{
"operation": "finish",
"time": "227.979",
"value": "finish"
}
]
}
|
21306
|
Pavla Kinclová, Daniela Bártová, and Lenka Oulehlová competed in the pole vault competition for Czechoslovakia at the 1992 Summer Olympics.
|
{
"evidence": [
"Pole vault at the Olympics_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:\nPavla Kinclová, Daniela Bártová, and Lenka Oulehlová competed in the pole vault competition for Czechoslovakia at the 1992 Summer Olympics.\n",
"role": "user"
}
] |
Search terms not in claim
|
{
"ground_truth": {
"evidence": [
"Pole vault at the Olympics_sentence_2"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 71,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0",
"value": "?search="
},
{
"operation": "search",
"time": "0.007",
"value": "Pavla Kinclová"
},
{
"operation": "Now on",
"time": "0.008",
"value": "Pavla Kinclová"
},
{
"operation": "search",
"time": "0.026",
"value": "1992 Summer Olympics"
},
{
"operation": "Now on",
"time": "0.027",
"value": "1992 Summer Olympics"
},
{
"operation": "Page search",
"time": "0.094",
"value": "pole"
},
{
"operation": "Highlighting",
"time": "0.151",
"value": "1992 Summer Olympics_cell_2_3_0"
},
{
"operation": "Highlighting",
"time": "0.151",
"value": "1992 Summer Olympics_title"
},
{
"operation": "Highlighting",
"time": "0.151",
"value": "1992 Summer Olympics_section_6"
},
{
"operation": "Highlighting",
"time": "0.151",
"value": "1992 Summer Olympics_header_cell_2_0_0"
},
{
"operation": "Highlighting deleted",
"time": "0.152",
"value": "1992 Summer Olympics_cell_2_3_0"
},
{
"operation": "Highlighting deleted",
"time": "0.152",
"value": "1992 Summer Olympics_title"
},
{
"operation": "Highlighting deleted",
"time": "0.152",
"value": "1992 Summer Olympics_section_6"
},
{
"operation": "Highlighting deleted",
"time": "0.152",
"value": "1992 Summer Olympics_header_cell_2_0_0"
},
{
"operation": "hyperlink",
"time": "0.153",
"value": "Athletics"
},
{
"operation": "Now on",
"time": "0.153",
"value": "Athletics at the 1992 Summer Olympics"
},
{
"operation": "hyperlink",
"time": "0.169",
"value": "Czechoslovakia"
},
{
"operation": "Now on",
"time": "0.17",
"value": "Czechoslovakia at the 1992 Summer Olympics"
},
{
"operation": "Page search",
"time": "0.197",
"value": "pole"
},
{
"operation": "hyperlink",
"time": "0.231",
"value": "Gymnastics at the 1992 Summer Olympics"
},
{
"operation": "Now on",
"time": "0.231",
"value": "Gymnastics at the 1992 Summer Olympics"
},
{
"operation": "Page search",
"time": "0.238",
"value": "pole"
},
{
"operation": "back-button-clicked",
"time": "0.245",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "0.246",
"value": "Czechoslovakia at the 1992 Summer Olympics"
},
{
"operation": "back-button-clicked",
"time": "0.253",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "0.253",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "0.253",
"value": "Athletics at the 1992 Summer Olympics"
},
{
"operation": "Page search",
"time": "0.265",
"value": "Pavla Kinclová"
},
{
"operation": "hyperlink",
"time": "0.285",
"value": "details"
},
{
"operation": "Now on",
"time": "0.286",
"value": "Athletics at the 1992 Summer Olympics – Men's pole vault"
},
{
"operation": "back-button-clicked",
"time": "0.288",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "0.288",
"value": "Athletics at the 1992 Summer Olympics"
},
{
"operation": "back-button-clicked",
"time": "0.324",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "0.324",
"value": "back-button-clicked"
},
{
"operation": "back-button-clicked",
"time": "0.324",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "0.324",
"value": "1992 Summer Olympics"
},
{
"operation": "hyperlink",
"time": "0.328",
"value": "Gymnastics"
},
{
"operation": "Now on",
"time": "0.328",
"value": "Gymnastics at the 1992 Summer Olympics"
},
{
"operation": "back-button-clicked",
"time": "0.35",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "0.35",
"value": "1992 Summer Olympics"
},
{
"operation": "hyperlink",
"time": "0.354",
"value": "Athletics"
},
{
"operation": "Now on",
"time": "0.354",
"value": "Athletics at the 1992 Summer Olympics"
},
{
"operation": "hyperlink",
"time": "0.365",
"value": "details"
},
{
"operation": "Now on",
"time": "0.366",
"value": "Athletics at the 1992 Summer Olympics – Men's pole vault"
},
{
"operation": "search",
"time": "0.394",
"value": "Contains...Athletics at the 1992 Summer Olympics – women's pole vault"
},
{
"operation": "Now on",
"time": "0.397",
"value": "undefined"
},
{
"operation": "back-button-clicked",
"time": "0.402",
"value": "back-button-clicked"
},
{
"operation": "Now on",
"time": "0.402",
"value": "Athletics at the 1992 Summer Olympics – Men's pole vault"
},
{
"operation": "Page search",
"time": "0.406",
"value": "Pavla"
},
{
"operation": "search",
"time": "0.418",
"value": "Daniela Bártová"
},
{
"operation": "Now on",
"time": "0.419",
"value": "Daniela Bártová"
},
{
"operation": "hyperlink",
"time": "0.476",
"value": "1992 Summer Olympics"
},
{
"operation": "Now on",
"time": "0.477",
"value": "1992 Summer Olympics"
},
{
"operation": "hyperlink",
"time": "0.596",
"value": "Athletics"
},
{
"operation": "Now on",
"time": "0.597",
"value": "Athletics at the 1992 Summer Olympics"
},
{
"operation": "search",
"time": "0.606",
"value": "Pole vault at the Olympics"
},
{
"operation": "Now on",
"time": "0.606",
"value": "Pole vault at the Olympics"
},
{
"operation": "Highlighting",
"time": "0.612",
"value": "Pole vault at the Olympics_sentence_2"
},
{
"operation": "finish",
"time": "0.669",
"value": "finish"
},
{
"operation": "Now on",
"time": "1619795757.467",
"value": "Pole vault at the Olympics"
},
{
"operation": "finish",
"time": "1619795760.53",
"value": "finish"
}
]
}
|
exfever_4453
|
Sammy Cahn is best known for his romantic lyrics to films and Broadway songs, as well as stand-alone songs premiered by recording companies in the second-largest metropolitan region in the United States.
|
{
"evidence": [
"Sammy Cahn",
"Greater Los Angeles Area"
],
"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:\nSammy Cahn is best known for his romantic lyrics to films and Broadway songs, as well as stand-alone songs premiered by recording companies in the second-largest metropolitan region in the United States.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Sammy Cahn",
"Greater Los Angeles Area"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 72,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
77796
|
Sadam Ali is an American professional boxer whose bouts are fought for a purse that is divided between the boxers as determined by contract.
|
{
"evidence": [
"Sadam Ali_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:\nSadam Ali is an American professional boxer whose bouts are fought for a purse that is divided between the boxers as determined by contract.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Sadam Ali_sentence_0"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 73,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.869",
"value": "?search="
},
{
"operation": "search",
"time": "5.836",
"value": "Sadam Ali"
},
{
"operation": "Now on",
"time": "7.584",
"value": "Sadam Ali"
},
{
"operation": "Highlighting",
"time": "10.661",
"value": "Sadam Ali_sentence_0"
},
{
"operation": "Page search",
"time": "38.594",
"value": "bout"
},
{
"operation": "Page search",
"time": "69.014",
"value": "purse"
},
{
"operation": "Page search",
"time": "79.438",
"value": "contract"
},
{
"operation": "finish",
"time": "127.112",
"value": "finish"
}
]
}
|
93862
|
The Northwest Film Center offers pre-school credit in association with pre-school of the United States, whereas the core program contains about ten classes covering fundamentals of video, film, and other aspects.
|
{
"evidence": [
"Northwest Film Center_sentence_9"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Northwest Film Center offers pre-school credit in association with pre-school of the United States, whereas the core program contains about ten classes covering fundamentals of video, film, and other aspects.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Northwest Film Center_sentence_9"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 74,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.118",
"value": "?search="
},
{
"operation": "search",
"time": "9.705",
"value": "The Northwest Film Center"
},
{
"operation": "Now on",
"time": "10.543",
"value": "Northwest Film Center"
},
{
"operation": "Page search",
"time": "52.826",
"value": "pre-school"
},
{
"operation": "Highlighting",
"time": "57.225",
"value": "Northwest Film Center_sentence_9"
},
{
"operation": "finish",
"time": "80.598",
"value": "finish"
}
]
}
|
89656
|
Tel Quel originally published essays and had an editors committee of 13 people; in 1971 it declared its support for Maoism.
|
{
"evidence": [
"Tel Quel_sentence_6",
"Tel Quel_sentence_7",
"Tel Quel_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:\nTel Quel originally published essays and had an editors committee of 13 people; in 1971 it declared its support for Maoism.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Tel Quel_sentence_6",
"Tel Quel_sentence_7",
"Tel Quel_sentence_9"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 75,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.194",
"value": "?search="
},
{
"operation": "search",
"time": "17.881",
"value": "Tel Quel"
},
{
"operation": "Now on",
"time": "18.943",
"value": "Tel Quel"
},
{
"operation": "Highlighting",
"time": "48.605",
"value": "Tel Quel_sentence_4"
},
{
"operation": "Highlighting deleted",
"time": "70.674",
"value": "Tel Quel_sentence_4"
},
{
"operation": "Highlighting",
"time": "71.412",
"value": "Tel Quel_sentence_6"
},
{
"operation": "Highlighting",
"time": "106.51",
"value": "Tel Quel_sentence_7"
},
{
"operation": "Highlighting",
"time": "113.346",
"value": "Tel Quel_sentence_5"
},
{
"operation": "Highlighting deleted",
"time": "114.823",
"value": "Tel Quel_sentence_5"
},
{
"operation": "Highlighting",
"time": "119.729",
"value": "Tel Quel_sentence_9"
},
{
"operation": "finish",
"time": "157.91",
"value": "finish"
}
]
}
|
exfever_4195
|
Hernani was Victor-Marie Hugo's first dramatic work.
|
{
"evidence": [
"Victor Hugo",
"Cromwell (play)",
"Romanticism"
],
"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:\nHernani was Victor-Marie Hugo's first dramatic work.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Victor Hugo",
"Cromwell (play)",
"Romanticism"
],
"label": "NOT ENOUGH INFO"
},
"style": "rule"
}
|
{
"index": 76,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
exfever_2979
|
Barbara Stanwyck began acting in talking pictures and starred in a 1930 American Pre-Code romantic drama film directed by Frank Capra.
|
{
"evidence": [
"Barbara Stanwyck",
"Ladies of Leisure"
],
"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:\nBarbara Stanwyck began acting in talking pictures and starred in a 1930 American Pre-Code romantic drama film directed by Frank Capra.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Barbara Stanwyck",
"Ladies of Leisure"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 77,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
1399
|
Although Piet Wildschut played for both FC Twente and PSV Eindhoven, he did not get the chance to represent his native country in the FIFA World Cup.
|
{
"evidence": [
"Piet Wildschut_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:\nAlthough Piet Wildschut played for both FC Twente and PSV Eindhoven, he did not get the chance to represent his native country in the FIFA World Cup.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Piet Wildschut_sentence_1"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 78,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "search",
"time": "56.598",
"value": "Piet Wildschut"
},
{
"operation": "Highlighting",
"time": "72.806",
"value": "Piet Wildschut_sentence_1"
},
{
"operation": "finish",
"time": "85.357",
"value": "finish"
}
]
}
|
77375
|
A Marie biscuit, also known as María, Mariebon, Мария and Marietta, is a type of biscuit similar to a rich tea biscuit created by the London bakery Peek Freans in 1874 which are frequently served to children.
|
{
"evidence": [
"Marie biscuit_sentence_0",
"Marie biscuit_sentence_1",
"Marie biscuit_sentence_4",
"Marie biscuit_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:\nA Marie biscuit, also known as María, Mariebon, Мария and Marietta, is a type of biscuit similar to a rich tea biscuit created by the London bakery Peek Freans in 1874 which are frequently served to children.\n",
"role": "user"
}
] |
Multi-hop Reasoning
|
{
"ground_truth": {
"evidence": [
"Marie biscuit_sentence_0",
"Marie biscuit_sentence_1",
"Marie biscuit_sentence_4",
"Marie biscuit_sentence_8"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 79,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.209",
"value": "?search="
},
{
"operation": "search",
"time": "22.762",
"value": "Marie biscuit"
},
{
"operation": "Now on",
"time": "23.972",
"value": "Marie biscuit"
},
{
"operation": "Highlighting",
"time": "25.998",
"value": "Marie biscuit_sentence_0"
},
{
"operation": "Highlighting",
"time": "33.72",
"value": "Marie biscuit_sentence_1"
},
{
"operation": "Highlighting",
"time": "50.995",
"value": "Marie biscuit_sentence_4"
},
{
"operation": "Page search",
"time": "73.09",
"value": "children"
},
{
"operation": "Page search",
"time": "73.409",
"value": "children"
},
{
"operation": "Page search",
"time": "73.649",
"value": "children"
},
{
"operation": "Page search",
"time": "73.864",
"value": "children"
},
{
"operation": "Highlighting",
"time": "78.619",
"value": "Marie biscuit_sentence_8"
},
{
"operation": "finish",
"time": "92.083",
"value": "finish"
}
]
}
|
23443
|
Andawala is a remote village in Central Province, Sri Lanka with two Buddhist temples and where most of the 400 families are working as farmers.
|
{
"evidence": [
"Andawala_sentence_0",
"Andawala_sentence_2",
"Andawala_sentence_8",
"Andawala_sentence_9",
"Andawala_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:\nAndawala is a remote village in Central Province, Sri Lanka with two Buddhist temples and where most of the 400 families are working as farmers.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"Andawala_sentence_0",
"Andawala_sentence_2",
"Andawala_sentence_8",
"Andawala_sentence_9",
"Andawala_sentence_5"
],
"label": "SUPPORT"
},
"style": "rule"
}
|
{
"index": 80,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.875",
"value": "?search="
},
{
"operation": "search",
"time": "10.284",
"value": "Andawala"
},
{
"operation": "Now on",
"time": "10.967",
"value": "Andawala"
},
{
"operation": "Highlighting",
"time": "18.628",
"value": "Andawala_sentence_2"
},
{
"operation": "Highlighting",
"time": "29.165",
"value": "Andawala_sentence_0"
},
{
"operation": "Highlighting deleted",
"time": "37.321",
"value": "Andawala_sentence_2"
},
{
"operation": "Highlighting",
"time": "38.207",
"value": "Andawala_sentence_1"
},
{
"operation": "Highlighting deleted",
"time": "39.114",
"value": "Andawala_sentence_1"
},
{
"operation": "Highlighting",
"time": "39.911",
"value": "Andawala_sentence_2"
},
{
"operation": "Page search",
"time": "53.141",
"value": "temple"
},
{
"operation": "Highlighting",
"time": "56.416",
"value": "Andawala_sentence_8"
},
{
"operation": "Highlighting",
"time": "62.062",
"value": "Andawala_sentence_9"
},
{
"operation": "Highlighting",
"time": "75.768",
"value": "Andawala_sentence_5"
},
{
"operation": "finish",
"time": "173.035",
"value": "finish"
}
]
}
|
exfever_3649
|
Naruto is a comics originating from Korea written and illustrated by Masashi Kishimoto that tells the story of Naruto Uzumaki.
|
{
"evidence": [
"Naruto",
"Manga"
],
"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:\nNaruto is a comics originating from Korea written and illustrated by Masashi Kishimoto that tells the story of Naruto Uzumaki.\n",
"role": "user"
}
] |
EX-FEVER
|
{
"ground_truth": {
"evidence": [
"Naruto",
"Manga"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 81,
"split": "train_online"
}
|
{
"annotator_operations": []
}
|
19001
|
The Cecil Kelley criticality accident involved an inexperienced 28-year-old Kelley who was permitted to operate a large capacity mixing tank of residual plutonium-239 and various organic solvents and acids.
|
{
"evidence": [
"Cecil Kelley criticality accident_sentence_3"
],
"label": "REFUTE"
}
|
feverous
|
[
{
"content": "You are a claim-verification assistant. You MUST follow this protocol exactly:\n\n<plan>…</plan>\n- Once at the start: sketch your high-level strategy, such as claim decomposition, entity recognition, etc.\n\n<search>…</search>\n- When you need a fact: emit exactly this tag with your query.\n- To make the most of your search turns, don’t repeat identical queries.\n- You can search at most three times.\n\n<information>\n[[e_1]]: info1\n[[e_2]]: info2\n…\n</information>\n- You will be given claim related information in the format above.\n\n<think>…</think> \n- Use for every piece of reasoning; do not state your final verdict here.\n- You must conduct reasoning inside <think> and </think> first every time you get new information.\n\n<answer>\nLabel: SUPPORT / REFUTE / NOT ENOUGH INFO\nEvidence: [[e_1]], [[e_3]], ...\n</answer> \n- Emit exactly once at the end, no extra text or tags.\n- Evidence id such as e_1 will be replaced by real ids from the corpus. You must include useful real ids when answering\n- Evidence outputs must strictly enforce the format [[e_i]], [[e_j]]…\n- Answer Labels respectively stand for:\nSUPPORT: The claim is consistent with the cited evidence and the evidence is sufficient to confirm the claim.\nREFUTE: The claim contradicts the cited evidence and the evidence is sufficient to disprove the claim.\nNOT ENOUGH INFO: The available evidence is insufficient to determine whether the claim is true or false.\n\n- Note that the process should be: plan → (search → information → think) repeat until conclusion → answer\n\nVerify the claim:\nThe Cecil Kelley criticality accident involved an inexperienced 28-year-old Kelley who was permitted to operate a large capacity mixing tank of residual plutonium-239 and various organic solvents and acids.\n",
"role": "user"
}
] |
Numerical Reasoning
|
{
"ground_truth": {
"evidence": [
"Cecil Kelley criticality accident_sentence_3"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 82,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "1.09",
"value": "?search="
},
{
"operation": "search",
"time": "26.474",
"value": "Cecil Kelley criticality accident"
},
{
"operation": "Now on",
"time": "27.287",
"value": "Cecil Kelley criticality accident"
},
{
"operation": "Page search",
"time": "127.772",
"value": "inexperienced"
},
{
"operation": "Highlighting",
"time": "202.145",
"value": "Cecil Kelley criticality accident_sentence_3"
},
{
"operation": "finish",
"time": "231.407",
"value": "finish"
}
]
}
|
41993
|
The 1980 Seattle Seahawks season was the team's fourth season in the National Football League, starting off at 4–3, then losing the remaining nine games of the season.
|
{
"evidence": [
"1980 Seattle Seahawks season_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 1980 Seattle Seahawks season was the team's fourth season in the National Football League, starting off at 4–3, then losing the remaining nine games of the season.\n",
"role": "user"
}
] |
Other
|
{
"ground_truth": {
"evidence": [
"1980 Seattle Seahawks season_sentence_0"
],
"label": "REFUTE"
},
"style": "rule"
}
|
{
"index": 83,
"split": "train_online"
}
|
{
"annotator_operations": [
{
"operation": "start",
"time": "0",
"value": "start"
},
{
"operation": "Now on",
"time": "0.828",
"value": "?search="
},
{
"operation": "search",
"time": "44.365",
"value": "1980 Seattle Seahawks season"
},
{
"operation": "Now on",
"time": "45.485",
"value": "1980 Seattle Seahawks season"
},
{
"operation": "Highlighting",
"time": "71.963",
"value": "1980 Seattle Seahawks season_sentence_0"
},
{
"operation": "finish",
"time": "110.654",
"value": "finish"
}
]
}
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.