Dataset Viewer
Auto-converted to Parquet
id
stringlengths
7
11
question
stringlengths
13
654
golden_answers
sequencelengths
1
25
data_source
stringclasses
2 values
prompt
listlengths
1
1
ability
stringclasses
1 value
reward_model
dict
extra_info
dict
metadata
dict
train_0
total number of death row inmates in the us?
[ "2,718" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: total number of death row inmates in the us?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "2,718" ] }, "style": "rule" }
{ "index": 0, "split": "train" }
null
train_1
big little lies season 2 how many episodes?
[ "seven" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: big little lies season 2 how many episodes?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "seven" ] }, "style": "rule" }
{ "index": 1, "split": "train" }
null
train_2
who sang waiting for a girl like you?
[ "Foreigner" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who sang waiting for a girl like you?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Foreigner" ] }, "style": "rule" }
{ "index": 2, "split": "train" }
null
train_3
where do you cross the arctic circle in norway?
[ "Saltfjellet" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: where do you cross the arctic circle in norway?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Saltfjellet" ] }, "style": "rule" }
{ "index": 3, "split": "train" }
null
train_4
who is the main character in green eggs and ham?
[ "Sam-I-am" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who is the main character in green eggs and ham?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Sam-I-am" ] }, "style": "rule" }
{ "index": 4, "split": "train" }
null
train_5
do veins carry blood to the heart or away?
[ "to" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: do veins carry blood to the heart or away?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "to" ] }, "style": "rule" }
{ "index": 5, "split": "train" }
null
train_6
who played charlie bucket in the original charlie and the chocolate factory?
[ "Peter Gardner[4] Ostrum" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who played charlie bucket in the original charlie and the chocolate factory?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Peter Gardner[4] Ostrum" ] }, "style": "rule" }
{ "index": 6, "split": "train" }
null
train_7
what is 1 radian in terms of pi?
[ "1/2π" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: what is 1 radian in terms of pi?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "1/2π" ] }, "style": "rule" }
{ "index": 7, "split": "train" }
null
train_8
when does season 5 of bates motel come out?
[ "February 20, 2017" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: when does season 5 of bates motel come out?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "February 20, 2017" ] }, "style": "rule" }
{ "index": 8, "split": "train" }
null
train_9
how many episodes are in series 7 game of thrones?
[ "seven" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: how many episodes are in series 7 game of thrones?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "seven" ] }, "style": "rule" }
{ "index": 9, "split": "train" }
null
train_10
who is next in line to be the monarch of england?
[ "Charles, Prince of Wales" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who is next in line to be the monarch of england?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Charles, Prince of Wales" ] }, "style": "rule" }
{ "index": 10, "split": "train" }
null
train_11
who is in charge of enforcing the pendleton act of 1883?
[ "United States Civil Service Commission" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who is in charge of enforcing the pendleton act of 1883?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "United States Civil Service Commission" ] }, "style": "rule" }
{ "index": 11, "split": "train" }
null
train_12
what is the name of latest version of android?
[ "Oreo" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: what is the name of latest version of android?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Oreo" ] }, "style": "rule" }
{ "index": 12, "split": "train" }
null
train_13
why was there so much interest in cuba both before and after the civil war?
[ "sugar markets" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: why was there so much interest in cuba both before and after the civil war?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "sugar markets" ] }, "style": "rule" }
{ "index": 13, "split": "train" }
null
train_14
when did veterans day start being called veterans day?
[ "June 1, 1954" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: when did veterans day start being called veterans day?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "June 1, 1954" ] }, "style": "rule" }
{ "index": 14, "split": "train" }
null
train_15
when did big air snowboarding become an olympic sport?
[ "2018" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: when did big air snowboarding become an olympic sport?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "2018" ] }, "style": "rule" }
{ "index": 15, "split": "train" }
null
train_16
who played in the most world series games?
[ "the New York Yankees" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who played in the most world series games?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "the New York Yankees" ] }, "style": "rule" }
{ "index": 16, "split": "train" }
null
train_17
who sings i can't stop this feeling anymore?
[ "American rock band REO Speedwagon" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who sings i can't stop this feeling anymore?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "American rock band REO Speedwagon" ] }, "style": "rule" }
{ "index": 17, "split": "train" }
null
train_18
who is the month of may named after?
[ "the Greek Goddess Maia" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who is the month of may named after?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "the Greek Goddess Maia" ] }, "style": "rule" }
{ "index": 18, "split": "train" }
null
train_19
who has the most petroleum in the world?
[ "Venezuela" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who has the most petroleum in the world?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Venezuela" ] }, "style": "rule" }
{ "index": 19, "split": "train" }
null
train_20
who is the sister of for king and country?
[ "Rebecca St. James" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who is the sister of for king and country?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Rebecca St. James" ] }, "style": "rule" }
{ "index": 20, "split": "train" }
null
train_21
who developed the first periodic table with 8 columns?
[ "Dmitri Mendeleev" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who developed the first periodic table with 8 columns?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Dmitri Mendeleev" ] }, "style": "rule" }
{ "index": 21, "split": "train" }
null
train_22
who plays skyler on lab rats elite force?
[ "Paris Berelc" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who plays skyler on lab rats elite force?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Paris Berelc" ] }, "style": "rule" }
{ "index": 22, "split": "train" }
null
train_23
when is season seven of game of thrones coming out?
[ "July 16, 2017" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: when is season seven of game of thrones coming out?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "July 16, 2017" ] }, "style": "rule" }
{ "index": 23, "split": "train" }
null
train_24
who went home on rupaul's drag race season 10 episode 4?
[ "Dusty Ray Bottoms" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who went home on rupaul's drag race season 10 episode 4?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Dusty Ray Bottoms" ] }, "style": "rule" }
{ "index": 24, "split": "train" }
null
train_25
when did nigeria become a federation of 19 states?
[ "1976" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: when did nigeria become a federation of 19 states?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "1976" ] }, "style": "rule" }
{ "index": 25, "split": "train" }
null
train_26
when did roller derby first appear in the press?
[ "1922" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: when did roller derby first appear in the press?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "1922" ] }, "style": "rule" }
{ "index": 26, "split": "train" }
null
train_27
what nba team has the highest winning percentage?
[ "San Antonio Spurs" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: what nba team has the highest winning percentage?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "San Antonio Spurs" ] }, "style": "rule" }
{ "index": 27, "split": "train" }
null
train_28
if there is a random change in the genetics of a small population it is termed?
[ "genetic drift" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: if there is a random change in the genetics of a small population it is termed?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "genetic drift" ] }, "style": "rule" }
{ "index": 28, "split": "train" }
null
train_29
who played vincent in nanny mcphee and the big bang?
[ "Oscar Steer" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who played vincent in nanny mcphee and the big bang?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Oscar Steer" ] }, "style": "rule" }
{ "index": 29, "split": "train" }
null
train_30
who played the eldest daughter in the sound of music?
[ "Charmian Carr" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who played the eldest daughter in the sound of music?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Charmian Carr" ] }, "style": "rule" }
{ "index": 30, "split": "train" }
null
train_31
who plays andy's teacher in parks and rec?
[ "Danielle Bisutti" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who plays andy's teacher in parks and rec?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Danielle Bisutti" ] }, "style": "rule" }
{ "index": 31, "split": "train" }
null
train_32
in which channel fifa world cup will be broadcasted?
[ "NBC", "Telemundo" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: in which channel fifa world cup will be broadcasted?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "NBC", "Telemundo" ] }, "style": "rule" }
{ "index": 32, "split": "train" }
null
train_33
what is a large body of water that flows through land?
[ "River" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: what is a large body of water that flows through land?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "River" ] }, "style": "rule" }
{ "index": 33, "split": "train" }
null
train_34
who played all the carly's on general hospital?
[ "Jennifer Bransford", "Tamara Braun", "Sarah Joy Brown", "Laura Wright" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who played all the carly's on general hospital?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Jennifer Bransford", "Tamara Braun", "Sarah Joy Brown", "Laura Wright" ] }, "style": "rule" }
{ "index": 34, "split": "train" }
null
train_35
an object that has mass and occupies space is known as?
[ "matter" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: an object that has mass and occupies space is known as?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "matter" ] }, "style": "rule" }
{ "index": 35, "split": "train" }
null
train_36
who sang the original blinded by the light?
[ "Bruce Springsteen" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who sang the original blinded by the light?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Bruce Springsteen" ] }, "style": "rule" }
{ "index": 36, "split": "train" }
null
train_37
who is the yellow rose of texas named after?
[ "Emily D. West" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who is the yellow rose of texas named after?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Emily D. West" ] }, "style": "rule" }
{ "index": 37, "split": "train" }
null
train_38
who sings the song i can see clearly now the rain is gone?
[ "Johnny Nash" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who sings the song i can see clearly now the rain is gone?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Johnny Nash" ] }, "style": "rule" }
{ "index": 38, "split": "train" }
null
train_39
who was the rfc editor until 1998 just provide the family name?
[ "Postel" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who was the rfc editor until 1998 just provide the family name?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Postel" ] }, "style": "rule" }
{ "index": 39, "split": "train" }
null
train_40
what is the number 1 sport in the usa?
[ "American football" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: what is the number 1 sport in the usa?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "American football" ] }, "style": "rule" }
{ "index": 40, "split": "train" }
null
train_41
when was the flight of the bumblebee written?
[ "in 1899–1900" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: when was the flight of the bumblebee written?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "in 1899–1900" ] }, "style": "rule" }
{ "index": 41, "split": "train" }
null
train_42
who plays cullen bohannon in hell on wheels?
[ "Anson Adams Mount IV" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who plays cullen bohannon in hell on wheels?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Anson Adams Mount IV" ] }, "style": "rule" }
{ "index": 42, "split": "train" }
null
train_43
when does the new pokemon movie come out?
[ "July 15, 2017" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: when does the new pokemon movie come out?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "July 15, 2017" ] }, "style": "rule" }
{ "index": 43, "split": "train" }
null
train_44
how many oar athletes are in the olympics?
[ "168" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: how many oar athletes are in the olympics?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "168" ] }, "style": "rule" }
{ "index": 44, "split": "train" }
null
train_45
where is the honda classic played this year?
[ "Palm Beach Gardens, Florida" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: where is the honda classic played this year?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Palm Beach Gardens, Florida" ] }, "style": "rule" }
{ "index": 45, "split": "train" }
null
train_46
who won the nrl grand final in 2015?
[ "North Queensland Cowboys" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who won the nrl grand final in 2015?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "North Queensland Cowboys" ] }, "style": "rule" }
{ "index": 46, "split": "train" }
null
train_47
where was charlie and the choclate factory filmed?
[ "Pinewood Studios in England" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: where was charlie and the choclate factory filmed?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Pinewood Studios in England" ] }, "style": "rule" }
{ "index": 47, "split": "train" }
null
train_48
who owns the rights to the song happy birthday?
[ "Warner/Chappell Music" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who owns the rights to the song happy birthday?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Warner/Chappell Music" ] }, "style": "rule" }
{ "index": 48, "split": "train" }
null
train_49
who sang the song every breath you take?
[ "The Police" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who sang the song every breath you take?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "The Police" ] }, "style": "rule" }
{ "index": 49, "split": "train" }
null
train_50
when did the british hand-over sovereignty of hong kong back to china?
[ "1 July 1997" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: when did the british hand-over sovereignty of hong kong back to china?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "1 July 1997" ] }, "style": "rule" }
{ "index": 50, "split": "train" }
null
train_51
the first life forms to appear on earth were?
[ "putative fossilized microorganisms" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: the first life forms to appear on earth were?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "putative fossilized microorganisms" ] }, "style": "rule" }
{ "index": 51, "split": "train" }
null
train_52
who played the robot on lost in space 2018?
[ "Brian Steele" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who played the robot on lost in space 2018?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Brian Steele" ] }, "style": "rule" }
{ "index": 52, "split": "train" }
null
train_53
who came up with britain's got talent?
[ "Simon Cowell" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who came up with britain's got talent?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Simon Cowell" ] }, "style": "rule" }
{ "index": 53, "split": "train" }
null
train_54
what year did my pet monster come out?
[ "1986" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: what year did my pet monster come out?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "1986" ] }, "style": "rule" }
{ "index": 54, "split": "train" }
null
train_55
who made the beavis and butthead theme song?
[ "Mike Judge" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who made the beavis and butthead theme song?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Mike Judge" ] }, "style": "rule" }
{ "index": 55, "split": "train" }
null
train_56
who has won the 2017 womens singles wimbledon tennis tournament?
[ "Garbiñe Muguruza" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who has won the 2017 womens singles wimbledon tennis tournament?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Garbiñe Muguruza" ] }, "style": "rule" }
{ "index": 56, "split": "train" }
null
train_57
when was the wheel first used in egypt?
[ "4th Dynasty" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: when was the wheel first used in egypt?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "4th Dynasty" ] }, "style": "rule" }
{ "index": 57, "split": "train" }
null
train_58
which type of organism produces food in an ecosystem?
[ "autotrophs" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: which type of organism produces food in an ecosystem?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "autotrophs" ] }, "style": "rule" }
{ "index": 58, "split": "train" }
null
train_59
who wrote the song for once in my life?
[ "Orlando Murden", "Ron Miller" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who wrote the song for once in my life?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Orlando Murden", "Ron Miller" ] }, "style": "rule" }
{ "index": 59, "split": "train" }
null
train_60
who sang i could not ask for more first?
[ "Edwin McCain" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who sang i could not ask for more first?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Edwin McCain" ] }, "style": "rule" }
{ "index": 60, "split": "train" }
null
train_61
who is the singer of waiting for love?
[ "Simon Aldred" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who is the singer of waiting for love?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Simon Aldred" ] }, "style": "rule" }
{ "index": 61, "split": "train" }
null
train_62
who recorded it's a man's world?
[ "James Brown" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who recorded it's a man's world?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "James Brown" ] }, "style": "rule" }
{ "index": 62, "split": "train" }
null
train_63
when does season 14 of grey's anatomy come out?
[ "September 28, 2017" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: when does season 14 of grey's anatomy come out?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "September 28, 2017" ] }, "style": "rule" }
{ "index": 63, "split": "train" }
null
train_64
when was the last time the detroit lions won a championship?
[ "1993" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: when was the last time the detroit lions won a championship?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "1993" ] }, "style": "rule" }
{ "index": 64, "split": "train" }
null
train_65
who played general chang in star trek 6?
[ "Christopher Plummer" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who played general chang in star trek 6?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Christopher Plummer" ] }, "style": "rule" }
{ "index": 65, "split": "train" }
null
train_66
which part of earth is covered with water?
[ "71%" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: which part of earth is covered with water?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "71%" ] }, "style": "rule" }
{ "index": 66, "split": "train" }
null
train_67
who's playing han solo in the han solo movie?
[ "Alden Ehrenreich" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who's playing han solo in the han solo movie?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Alden Ehrenreich" ] }, "style": "rule" }
{ "index": 67, "split": "train" }
null
train_68
where did the rule of 72 come from?
[ "of Luca Pacioli", "Summa de arithmetica" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: where did the rule of 72 come from?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "of Luca Pacioli", "Summa de arithmetica" ] }, "style": "rule" }
{ "index": 68, "split": "train" }
null
train_69
when was the masque of the red death written?
[ "1842" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: when was the masque of the red death written?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "1842" ] }, "style": "rule" }
{ "index": 69, "split": "train" }
null
train_70
who sings the original version of feeling good?
[ "Cy Grant" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who sings the original version of feeling good?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Cy Grant" ] }, "style": "rule" }
{ "index": 70, "split": "train" }
null
train_71
how many fleet of war ships brought by achilles in the shore near troy?
[ "50" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: how many fleet of war ships brought by achilles in the shore near troy?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "50" ] }, "style": "rule" }
{ "index": 71, "split": "train" }
null
train_72
who does black panther marry in the comics?
[ "Ororo Munroe" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who does black panther marry in the comics?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Ororo Munroe" ] }, "style": "rule" }
{ "index": 72, "split": "train" }
null
train_73
what is the record for home runs in a season?
[ "Barry Bonds" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: what is the record for home runs in a season?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Barry Bonds" ] }, "style": "rule" }
{ "index": 73, "split": "train" }
null
train_74
how do you say horse meat in french?
[ "viande chevaline" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: how do you say horse meat in french?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "viande chevaline" ] }, "style": "rule" }
{ "index": 74, "split": "train" }
null
train_75
who is the chief law officer of the government of india?
[ "The Attorney General for India" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who is the chief law officer of the government of india?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "The Attorney General for India" ] }, "style": "rule" }
{ "index": 75, "split": "train" }
null
train_76
when did the original flatliners movie come out?
[ "1990" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: when did the original flatliners movie come out?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "1990" ] }, "style": "rule" }
{ "index": 76, "split": "train" }
null
train_77
who is bright eyes in the loved ones?
[ "Anne Scott-Pendlebury" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who is bright eyes in the loved ones?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Anne Scott-Pendlebury" ] }, "style": "rule" }
{ "index": 77, "split": "train" }
null
train_78
what was the geeks name in 16 candles?
[ "Anthony Michael Hall" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: what was the geeks name in 16 candles?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Anthony Michael Hall" ] }, "style": "rule" }
{ "index": 78, "split": "train" }
null
train_79
who has scored the most own goals in premier league history?
[ "Alan Shearer" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who has scored the most own goals in premier league history?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Alan Shearer" ] }, "style": "rule" }
{ "index": 79, "split": "train" }
null
train_80
what day of the year is mothers day always celebrated?
[ "the second Sunday in May" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: what day of the year is mothers day always celebrated?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "the second Sunday in May" ] }, "style": "rule" }
{ "index": 80, "split": "train" }
null
train_81
when was the first episode of only fools and horses made?
[ "8 September 1981" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: when was the first episode of only fools and horses made?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "8 September 1981" ] }, "style": "rule" }
{ "index": 81, "split": "train" }
null
train_82
what was the first bring it on movie?
[ "Bring It On" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: what was the first bring it on movie?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Bring It On" ] }, "style": "rule" }
{ "index": 82, "split": "train" }
null
train_83
when did the words in god we trust appear on money?
[ "1864" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: when did the words in god we trust appear on money?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "1864" ] }, "style": "rule" }
{ "index": 83, "split": "train" }
null
train_84
who plays smithys sister in gavin and stacey?
[ "Sheridan Smith" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who plays smithys sister in gavin and stacey?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Sheridan Smith" ] }, "style": "rule" }
{ "index": 84, "split": "train" }
null
train_85
who is the nets all time leading scorer?
[ "Brook Lopez" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who is the nets all time leading scorer?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Brook Lopez" ] }, "style": "rule" }
{ "index": 85, "split": "train" }
null
train_86
who sang you won't get me i'm part of the union?
[ "Strawbs" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who sang you won't get me i'm part of the union?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Strawbs" ] }, "style": "rule" }
{ "index": 86, "split": "train" }
null
train_87
where was the first pan african conference held?
[ "in London" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: where was the first pan african conference held?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "in London" ] }, "style": "rule" }
{ "index": 87, "split": "train" }
null
train_88
who played adam schiff on law & order?
[ "Steven Hill" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who played adam schiff on law & order?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Steven Hill" ] }, "style": "rule" }
{ "index": 88, "split": "train" }
null
train_89
who was president when the white house was built?
[ "John Adams" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who was president when the white house was built?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "John Adams" ] }, "style": "rule" }
{ "index": 89, "split": "train" }
null
train_90
where is the big statue of jesus located?
[ "Rio de Janeiro, Brazil" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: where is the big statue of jesus located?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Rio de Janeiro, Brazil" ] }, "style": "rule" }
{ "index": 90, "split": "train" }
null
train_91
who sang the song you are the reason?
[ "Calum Scott" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: who sang the song you are the reason?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Calum Scott" ] }, "style": "rule" }
{ "index": 91, "split": "train" }
null
train_92
what is the predominant religion in the ukraine?
[ "Orthodoxy" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: what is the predominant religion in the ukraine?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Orthodoxy" ] }, "style": "rule" }
{ "index": 92, "split": "train" }
null
train_93
when did the who perform at the super bowl?
[ "Feb 7, 2010" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: when did the who perform at the super bowl?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Feb 7, 2010" ] }, "style": "rule" }
{ "index": 93, "split": "train" }
null
train_94
where was practical magic supposed to be set?
[ "Maria's Island in Massachusetts" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: where was practical magic supposed to be set?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Maria's Island in Massachusetts" ] }, "style": "rule" }
{ "index": 94, "split": "train" }
null
train_95
when was abbott and costello who's on first?
[ "February 1938" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: when was abbott and costello who's on first?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "February 1938" ] }, "style": "rule" }
{ "index": 95, "split": "train" }
null
train_96
india's medal ranking in asian games 2018?
[ "8" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: india's medal ranking in asian games 2018?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "8" ] }, "style": "rule" }
{ "index": 96, "split": "train" }
null
train_97
how far is murrysville pa from pittsburgh pa?
[ "roughly 20 miles" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: how far is murrysville pa from pittsburgh pa?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "roughly 20 miles" ] }, "style": "rule" }
{ "index": 97, "split": "train" }
null
train_98
what is the pattern on a leopard's coat called?
[ "A leopard pattern" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: what is the pattern on a leopard's coat called?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "A leopard pattern" ] }, "style": "rule" }
{ "index": 98, "split": "train" }
null
train_99
where is eden west resort from couples retreat?
[ "Bora Bora" ]
nq
[ { "content": "Answer the given question. You must conduct reasoning inside <think> and </think> first every time you get new information. After reasoning, if you find you lack some knowledge, you can call a search engine by <search> query </search> and it will return the top searched results between <information> and </information>. You can search as many times as your want. If you find no further external knowledge needed, you can directly provide the answer inside <answer> and </answer>, without detailed illustrations. For example, <answer> Beijing </answer>. Question: where is eden west resort from couples retreat?\n", "role": "user" } ]
fact-reasoning
{ "ground_truth": { "target": [ "Bora Bora" ] }, "style": "rule" }
{ "index": 99, "split": "train" }
null
End of preview. Expand in Data Studio

No dataset card yet

Downloads last month
39

Collection including PeterJinGo/nq_hotpotqa_train