inputs stringlengths 38 313k | targets stringlengths 0 4.86k | _template_idx int64 0 9 | _task_source stringclasses 1
value | _task_name stringlengths 19 85 | _template_type stringclasses 2
values | embedding listlengths 1.02k 1.02k |
|---|---|---|---|---|---|---|
instruction:
In this task, you are given a question in English, and your task is to translate it into Persian.
question:
Can we expect time travel to become a reality?
answer:
آیا می توانیم انتظار داشته باشیم که سفر زمانی به واقعیت تبدیل شود؟
question:
Why do people often ask questions in Quora while they can Google it themselves?
answer:
چرا افراد غالباً در Quora سؤال می کنند ، در حالی که می توانند سوالات خود را در Google جستجو کنند؟
question:
What is life like after clearing SSC CGL?
answer:
| زندگی پس از پاس کردن SSC CGL چگونه است؟
| 9 | NIv2 | task652_parsinlu_en_fa_translation | fs_opt | [
-1.2500358819961548,
0.7001782655715942,
-0.2816048264503479,
-0.20671208202838898,
-0.7795726656913757,
-0.6031984090805054,
0.4873809218406677,
0.14791271090507507,
-0.07147908210754395,
-0.4341585040092468,
0.04241595417261124,
1.0617866516113281,
-0.10220178216695786,
0.270257622003555... |
Teacher: Given a command in a limited form of natural language, provide the correct sequence of actions that executes the command to thus navigate an agent in its environment. A command can be broken down into many different actions. Actions are uppercase and are individual steps that serve as the building blocks for a command. For commands, 'left' and 'right' are used to denote the direction of an action. The word 'opposite' turns the agent backward in the specified direction. The word 'around' makes the agent execute an action while turning around in the specified direction. The word 'and' means to execute the next scope of the command following the previous scope of the command. The word 'after' signifies to execute the previous scope of the command following the next scope of the command. The words 'twice' and 'thrice' trigger repetition of a command that they scope over two times or three times, respectively. There are only six actions: 'I_LOOK', 'I_WALK', 'I_RUN', 'I_JUMP', 'I_TURN_LEFT', and 'I_TURN_RIGHT'. These actions respectively align with the commands 'look', 'walk', 'run', 'jump', 'turn left', and 'turn right'. Actions and commands do not have quotations in the input and output.
Teacher: Now, understand the problem? If you are still confused, see the following example:
jump left
Solution: I_TURN_LEFT I_JUMP
Reason: The agent must first turn left and then jump in order to jump to the left.
Now, solve this instance: run after look opposite right thrice
Student: | I_TURN_RIGHT I_TURN_RIGHT I_LOOK I_TURN_RIGHT I_TURN_RIGHT I_LOOK I_TURN_RIGHT I_TURN_RIGHT I_LOOK I_RUN | 2 | NIv2 | task128_scan_structured_text_generation_command_action_short | fs_opt | [
0.06278233230113983,
0.6610661745071411,
-0.1754833608865738,
-0.08289958536624908,
0.12633459270000458,
0.16430412232875824,
0.09993194043636322,
0.24077101051807404,
-0.17538048326969147,
-0.4052457809448242,
-0.7412257790565491,
-0.2688579559326172,
-0.7998842597007751,
-0.2036481499671... |
Given a set of five words, generate the word from the set that does not belong (i.e. is the least relevant) with the other words. Words are separated by commas.
One example is below.
Q: excavator, crane, pelican, hoist, upraise
A: pelican
Rationale: The common category here are words related to construction. A pelican is not related to construction because it is a type of bird.
Q: blind, deck, pack, jab, draw
A: | jab | 9 | NIv2 | task142_odd-man-out_classification_no_category | fs_opt | [
0.1483149230480194,
0.6266913414001465,
0.11680537462234497,
-0.06338660418987274,
0.01565648801624775,
0.20327132940292358,
0.45943140983581543,
0.17704996466636658,
0.046633280813694,
-0.2694254219532013,
-1.0177031755447388,
0.19611820578575134,
-0.6478044986724854,
0.39766231179237366,... |
In this task, you will be presented with an English sentence, and you have to translate it into Arabic.
Let me give you an example: But if the shares belong to Geoffrey Charles, well, then George cannot touch them.
The answer to this example can be: لكن لو كانت الأسهم بإسم جوفري تشارلز لن يمكن لجورج المساس بهم.
Here is why: This is a good example. The above sentence is correctly translated from English to Arabic.
OK. solve this:
Now, as you can clearly see, Bluebell is truly
Answer: | الآن كما ترون بكل وضوح بلوبيل | 8 | NIv2 | task651_opus100_en_ar_translation | fs_opt | [
0.15297219157218933,
0.767907977104187,
0.02612079679965973,
-1.0578746795654297,
-0.5443700551986694,
-0.10110408067703247,
1.1459614038467407,
0.28166982531547546,
0.21847708523273468,
-0.7191234230995178,
-0.4268275797367096,
0.6592732667922974,
-0.43133828043937683,
-0.1814582496881485... |
Instructions: In this task, you are given a sentence in the Hindi language and a corresponding English translation of the Hindi sentence. Your task is to generate a label "Yes" if the translation is correct, otherwise generate label "No". In the translation, English sentence should preserve the number as it is and it should be in sentence case (capitalize only the first word of each sentence and noun).
Input: Hindi: निश्चित करें स्कूल इस बात से अवगत है कि आप चाहतें हैं कि किसी भी समस्या के उत्पऋ होने पर , ऋसमें कि आपका बच्चा शामिल है आपको तुरंत ही इस बारे में बताया जाए |भाष्;
English: Make sure the school knows that you want to be told as soon as any problems arise which involve your child .
Output: | Yes | 3 | NIv2 | task426_hindienglish_corpora_hi-en_classification | zs_opt | [
-0.32960695028305054,
0.3140872120857239,
0.8074917793273926,
0.4161742329597473,
-0.15166769921779633,
-0.4736400246620178,
0.07094943523406982,
0.8280526399612427,
-0.2815873920917511,
-0.2600606679916382,
0.09731680154800415,
0.4829038381576538,
-0.8409922122955322,
-0.17565730214118958... |
Teacher: In this task, you need to answer 'Yes' if the given word is the longest word (in terms of number of letters) in the given sentence, else answer 'No'. Note that there could be multiple longest words in a sentence as they can have the same length that is the largest across all words in that sentence.
Teacher: Now, understand the problem? If you are still confused, see the following example:
Sentence: 'two men in a room holding wii remotes'. Is 'remotes' the longest word in the sentence?
Solution: Yes
Reason: The word 'remotes' has 7 letters which is the maximum in this sentence. So, the answer is 'Yes'.
Now, solve this instance: Sentence: 'a man rides a wave lying on a board'. Is 'a' the longest word in the sentence?
Student: | No | 2 | NIv2 | task114_is_the_given_word_longest | fs_opt | [
0.10001392662525177,
1.052027940750122,
-0.4669562578201294,
-0.5264261960983276,
-0.06135587766766548,
-0.638098955154419,
0.47877585887908936,
0.4924169182777405,
0.45959776639938354,
-0.5058243274688721,
-0.48606809973716736,
0.697984516620636,
-0.2474658489227295,
-0.3713342845439911,
... |
Teacher:In this task, you are given two phrases: Head and Tail, separated with <sep>. The Head and the Tail events are short phrases possibly involving participants. The names of specific people have been replaced by generic words (e.g., PersonX, PersonY, PersonZ). PersonX is always the subject of the event. You have to determine whether the Head is located or can be found at/in/on the Tail or not. Classify your answers into "Yes" and "No". The phrase may also contain "___", a placeholder that can be an object, a person, and/or an action.
Teacher: Now, understand the problem? Solve this instance: Head: cat<sep>Tail: hiding
Student: | Yes | 6 | NIv2 | task1207_atomic_classification_atlocation | zs_opt | [
0.03839327022433281,
0.19267407059669495,
0.2995471954345703,
0.18810614943504333,
-0.5349806547164917,
-0.9006046652793884,
0.7234489321708679,
0.3797191381454468,
-0.6418306827545166,
-0.49912261962890625,
-0.3898298144340515,
-0.37907058000564575,
-0.6489580273628235,
0.1908049583435058... |
In this task, you are given an app review. Your task is to identify whether its sentiment is "Positive" or "Negative".
Example Input: Really cool gadget for RF engineers Restarts driver when changing orientation. Please fix that. Additionally would like to have MUTE button alongside with other buttons. Another issue is: let user choose whether he/she wants kHz MHz or GHz when entering freq. Thanks in advance!
Example Output: Positive
Example Input: Loved it ! A single issue. Do not read my messages application Notify My Android.
Example Output: Positive
Example Input: doesn't support oneplus one
Example Output: | Negative
| 3 | NIv2 | task761_app_review_classification | fs_opt | [
-0.7163127660751343,
-0.18524664640426636,
-0.2279239147901535,
-0.8258809447288513,
0.3198297619819641,
-1.0915076732635498,
0.33781763911247253,
0.5904147028923035,
-0.12778450548648834,
0.25186118483543396,
-0.9092533588409424,
-0.5326523780822754,
0.4437803030014038,
-0.545697450637817... |
Instructions: Given a sentence in the Japanese and Thai language. Your task is check if the Filipino sentence is translation of Japanese. if the translation is correct than generate label "Yes", otherwise generate label "No".
Input: Japanese: 彼女は、ベトナムの教会から来た仲間の牧師によって慰められていた。
Thai: เขาได้ถูกนำตัวส่งโรงพยาบาลเมื่อวาน เพื่อเข้ารับการรักษาโรคปวดบวม จนกระทั่งเสียชีวิตเวลาประมาณ 1.45 AM (6.45 AM GMT)
Output: | No | 3 | NIv2 | task1129_alt_ja_th_answer_generation | zs_opt | [
0.04692881554365158,
-0.2650810182094574,
0.09695085883140564,
0.37266993522644043,
0.11287970840930939,
-1.4399739503860474,
0.772611141204834,
0.32640552520751953,
-0.2863563895225525,
-0.2897722125053406,
-0.21937210857868195,
0.6361725330352783,
0.017407236620783806,
0.8680626153945923... |
Detailed Instructions: In this task, you are given a short story consisting of exactly 5 sentences where the second sentence is missing. You are given two options and you need to select the one that best connects the first sentence with the rest of the story. Indicate your answer by 'Option 1' if the first option is correct, otherwise 'Option 2'. The incorrect option will change the subsequent storyline, so that at least one of the three subsequent sentences is no longer consistent with the story.
Q: Sentence 1: Elle and Ed were on a road trip.
Sentence 3: He scrutinized the map helplessly
Sentence 4: Then Elle took it and turned it over
Sentence 5: She was able to find their way back to the main road!
Option 1: Ed was lost on a small tertiary road.
Option 2: Elle was lost on a small road.
A: | Option 1 | 9 | NIv2 | task065_timetravel_consistent_sentence_classification | zs_opt | [
-0.750388503074646,
0.5934460759162903,
0.3450857400894165,
-0.27167296409606934,
0.1593950390815735,
-0.2926682233810425,
0.8097888231277466,
1.115452766418457,
-0.06907384097576141,
0.48007380962371826,
-0.05142861604690552,
-0.8098973035812378,
-0.09915803372859955,
-0.5261091589927673,... |
Given an abstract, indicate which of the provided candidate titles (Option1 and Option2) best describes it (respond with the text of the title, not option1/2).
[Q]: Abstract: We examined the effectiveness of instruments that describe a child's locus of control and perceived self-competence as predictors of outcome in the treatment of encopresis. Fifty children 9 to 12 years of age completed such instruments prior to clinical interview. Clinicians were blind as to the results of these questionnaires and administered a standard encopresis protocol to all patients. Six to 17 months later, 45 (90%) of the patient's parents were interviewed to assess compliance and treatment outcome. Children who significantly improved had a more internal locus of control (P less than 0.05) and better compliance (P less than 0.05) as determined by two separate outcome variables. Locus of control and compliance were not, however, significantly correlated with each other; each appeared to have an independent association with outcome. There was no association between perceived self-competence and outcome. The positive association between locus of control and outcome was not mediated by socioeconomic status, age, duration of encopresis, severity of encopresis, gender, or time between initiation of treatment and follow-up. Measurement of locus of control in children with encopresis appears to be helpful in identifing a group at high risk for resistance of treatment.
Which is the apt title for the abstract? option1: Locus of control as predictor of compliance and outcome in treatment of encopresis. option2: Light and electron microscopic findings with permanent eyeliner.
[A]: Locus of control as predictor of compliance and outcome in treatment of encopresis.
[Q]: Abstract: An exceptional form of hepatic alveolar echinococcosis with metastasis of the right atrium is reported. This cardiac location of the parasitosis was revealed by attacks of pulmonary embolism which produced secondary pulmonary lesions. This case suggests that pulmonary metastases of alveolar echinococcosis of the liver might be due to the migration of parasitic clots from the hepatic veins.
Which is the apt title for the abstract? option1: Pulmonary embolism from a parasitic cardiac clot secondary to hepatic alveolar echinococcosis. option2: Sphenoid sinus mucocele masquerading as a skull base malignancy.
[A]: Pulmonary embolism from a parasitic cardiac clot secondary to hepatic alveolar echinococcosis.
[Q]: Abstract: We compared the degree of complement activation in 15 patients receiving massive blood transfusions and 21 patients with the septic syndrome and examined its association with other risk factors for adult respiratory distress syndrome (ARDS). Patients with massive transfusion (n = 8) as their only risk factor for ARDS had lower C3a levels (250 +/- 80 ng/ml) and a lower incidence of ARDS (0%) than patients with massive transfusion plus other risk factors (n = 7; C3a, 600 +/- 120 ng/ml; ARDS, 55%) or patients with the septic syndrome (n = 21; C3a, 540 +/- 80 ng/ml; ARDS, 19%). In our patients, the degree of complement activation appeared to reflect the presence of complicating clinical conditions.
Which is the apt title for the abstract? option1: C3a and adult respiratory distress syndrome after massive transfusion. option2: Subcapsular hematoma in fatty liver of pregnancy.
[A]: | C3a and adult respiratory distress syndrome after massive transfusion.
| 5 | NIv2 | task624_ohsumed_question_answering | fs_opt | [
0.479388028383255,
0.11421291530132294,
-0.5142441987991333,
0.2053186595439911,
0.4785829186439514,
-0.722376823425293,
0.43180182576179504,
0.6871802806854248,
0.30738943815231323,
0.40358659625053406,
-0.43996500968933105,
0.30009591579437256,
-0.07230452448129654,
0.22583448886871338,
... |
You are given a paragraph, a question and some answer options (associated with "A", "B", "C", "D", etc.) You need to list all the correct answer options with their associated letters (e.g. "A" can be a valid answer). Note that sometimes, more than one option can answer the question correctly and completely. In those cases, please generate all such options separated by a comma (e.g. "A, B" is a valid answer). Do not generate anything else apart from one or more of the following characters: 'A', 'B, 'C', 'D', etc.
Q: Paragraph- Sent 1: There have been many organisms that have lived in Earths past.
Sent 2: Only a tiny number of them became fossils.
Sent 3: Still, scientists learn a lot from fossils.
Sent 4: Fossils are our best clues about the history of life on Earth.
Sent 5: Fossils provide evidence about life on Earth.
Sent 6: They tell us that life on Earth has changed over time.
Sent 7: Fossils in younger rocks look like animals and plants that are living today.
Sent 8: Fossils in older rocks are less like living organisms.
Sent 9: Fossils can tell us about where the organism lived.
Sent 10: Was it land or marine?
Sent 11: Fossils can even tell us if the water was shallow or deep.
Sent 12: Fossils can even provide clues to ancient climates.
Question: Name two things fossils can tell us about the enviroment? (A) What animals ate, why they died (B) Ancient climates, how organisms lived (C) They can reveal if it was made in shallow or deep water and climate issues (D) If man existed (E) How many animals there were (F) Tell us if the water was shallow or deep (G) What animals were extinct (H) Clues to ancient climates.
A: B, C, F, H.
****
Q: Paragraph- Sent 1: Once upon a time, there was a squirrel named Joey.
Sent 2: Joey loved to go outside and play with his cousin Jimmy.
Sent 3: Joey and Jimmy played silly games together, and were always laughing.
Sent 4: One day, Joey and Jimmy went swimming together at their Aunt Julie's pond.
Sent 5: Joey woke up early in the morning to eat some food before they left.
Sent 6: He couldn't find anything to eat except for pie!
Sent 7: Usually, Joey would eat cereal, fruit (a pear), or oatmeal for breakfast.
Sent 8: After he ate, he and Jimmy went to the pond.
Sent 9: On their way there they saw their friend Jack Rabbit.
Sent 10: They dove into the water and swam for several hours.
Sent 11: The sun was out, but the breeze was cold.
Sent 12: Joey and Jimmy got out of the water and started walking home.
Sent 13: Their fur was wet, and the breeze chilled them.
Sent 14: When they got home, they dried off, and Jimmy put on his favorite purple shirt.
Sent 15: Joey put on a blue shirt with red and green dots.
Sent 16: The two squirrels ate some food that Joey's mom, Jasmine, made and went off to bed.
Question: How did Joey and Jimmy spend their time together? (A) They spent their time laughing, playing silly games, playing outside at Aunt Julie's pond (B) Causing trouble (C) Walking swimming.
A: A, C.
****
Q: Paragraph- Sent 1: In Meadowvale , California , on June 9 , 1970 , three babies are born at the height of a total eclipse .
Sent 2: Due to the sun and moon blocking Saturn , which controls emotions , the babies develop into children who are heartless and uncaring , with no feelings of remorse for their awful offences .
Sent 3: Ten years later , in 1980 , Curtis , Debbie and Steven begin to kill the adults around them and are able to escape detection because of their youthful and innocent fa ades .
Sent 4: However , when Timmy and Joyce discover their crimes , they find themselves as the targets of the evil children .
Sent 5: Their attempt to warn their neighbors about them are ignored .
Sent 6: Soon enough , Beverly discovers Debbie 's horrible secret with the newspaper clippings as evidence as proof and learns that she , Curtis and Steven were responsible for the murders .
Sent 7: She realizes that Joyce and Timmy were telling the truth all along and confronts her sister for it .
Sent 8: Beverly and Mrs. Brody forbids Debbie from hanging out with Curtis and Steven .
Sent 9: Debbie responds by killing her older sister in retaliation with a bow and arrow .
Sent 10: They soon attempt to murder Joyce and Timmy .
Sent 11: However , they fight back and manages to capture Steven and Curtis .
Sent 12: While Timmy calls the police from a neighbor 's house , Debbie manages to escape from the backyard and gets inside her mother 's car .
Sent 13: She manages to lie about Curits and Steven trying to set her up to take the fall for the attacks .
Sent 14: At the end , Curtis and Steven are arrested by the police much to the shock and anger of the town .
Question: Who captures Steven and Curtis? (A) Dabbie and brody (B) Joyce and Timmy.
A: | B.
****
| 4 | NIv2 | task058_multirc_question_answering | fs_opt | [
0.2828262448310852,
0.41088032722473145,
-0.7663681507110596,
0.07093528658151627,
-0.15235857665538788,
-1.5269904136657715,
0.4099389314651489,
0.8228427171707153,
-0.375355064868927,
0.0006888438947498798,
-0.5902371406555176,
0.19638782739639282,
-0.5086725950241089,
0.1557908356189727... |
Teacher:In this task you are given a small conversation between two persons and 4 options on how the conversation should continue. Your job is to choose the most reasonable option. The conversation and the options are separated by a newline character. Each dialogue in the conversation are separated by a comma. F and M indicate female and male, respectively.
Teacher: Now, understand the problem? Solve this instance: F: I left school late and I was trying to get home before the rush hour. ,M: Did you make it? ,F: No. When the police officer was giving me a ticket the rush hour started. It took me twice as long to get home.
(A) M: Oh, you want to get to the work place in a hurry to avoid the rush hour. (B) M: I am sorry. You spoke so rapidly that I did not catch the meaning of what you said. Could you repeat that? (C) M: Why do you want to avoid a ticket? (D) M: To get rid of the police is ridiculous.
Student: | B | 6 | NIv2 | task611_mutual_multi_turn_dialogue | zs_opt | [
-0.2244722694158554,
0.2650187611579895,
0.2467242181301117,
-0.340623676776886,
0.41734805703163147,
0.4295951724052429,
0.45566749572753906,
0.7412590980529785,
0.4419955015182495,
-0.1255444437265396,
-0.6420129537582397,
-0.11900864541530609,
0.18376097083091736,
-0.6021186709403992,
... |
Detailed Instructions: The provided file includes inquiries about restaurants, and we ask you to translate those to the Polish language. Please bear in mind the following guidlines while doing the translation: 1) We are looking for the most naturally written and formal form of each sentence in your language. We are *NOT* looking for colloquial forms of the sentence. We are looking for formal form which is how you would type your queries in a text-based virtual assistant. 2) The words between quotation marks *SHOULD NOT* be translated. We expect you to keep those values intact and include the quotation marks around them as well. 3) The fully capitalized words like DATE_0, or DURATION_0 *SHOULD NOT* be translated. Please keep them as they are in the translations. 4) Please do not localize measurement units like miles to kilometers during your translation. miles should be translated to its equivalent in your language. 6) Note the input is all lowercased except for special placeholders. Please do the same in your translations.
See one example below:
Problem: are there any " italian " restaurants nearby with 3 star reviews ?
Solution: czy w pobliżu znajdują się jakiekolwiek restauracje " italian " z opiniami z 3 gwiazdką/gwiazdkami?
Explanation: The translation correctly preserves " italian " entity and is accurate
Problem: i would like to find " brenda 's french soul food " .
Solution: | chcę znaleźć "brenda 's french soul food". | 4 | NIv2 | task175_spl_translation_en_pl | fs_opt | [
-0.5226274728775024,
0.8451075553894043,
-0.2535952627658844,
0.14051499962806702,
-0.05936259776353836,
-0.5543856620788574,
1.1362829208374023,
0.18568947911262512,
-0.2044421136379242,
-0.35261765122413635,
0.2222907841205597,
0.3428923487663269,
-0.39898496866226196,
0.2532093524932861... |
Detailed Instructions: In this task, you are given a context, a subject, a relation, and many options. Based on the context, from the options select the object entity that has the given relation with the subject. Answer with text (not indexes).
See one example below:
Problem: Context: Joanne McLeod is a Canadian figure skating coach. She is the skating director at the Champs International Skating Centre of BC (formerly known as the BC Centre of Excellence). Here current and former students include Emanuel Sandhu, Mira Leung, Kevin Reynolds, Jeremy Ten, Nam Nguyen, and many others. In 2012, McLeod became the first level 5 certified figure skating coach in British Columbia., Victor Kraatz, MSC (born April 7, 1971) is a Canadian former ice dancer. In 2003, he and his partner, Shae-Lynn Bourne, became the first North American ice dancers to win a World Championship., Allie Hann-McCurdy (born May 23, 1987 in Nanaimo, British Columbia) is a Canadian ice dancer. McCurdy began skating at age eight and was a singles skater until age 12 when she switched to ice dancing. In 2003 she teamed up with Michael Coreno, with whom she was the 2010 Four Continents silver medalist and the 2008 Canadian bronze medalist. The pair retired in June 2010, to coach at the Gloucester Skating Club., Maikki Uotila - Kraatz ( born 25 February 1977 ) is a Finnish ice dancer . She is a former Finnish national champion with Toni Mattila . She married Victor Kraatz on June 19 , 2004 . The two coach in Vancouver , where they are the ice dancing directors at the BC Centre of Excellence . She and Kraatz have two sons , born September 14 , 2006 and July 10 , 2010 ., Burnaby is a city in British Columbia, Canada, located immediately to the east of Vancouver. It is the third-largest city in British Columbia by population, surpassed only by nearby Surrey and Vancouver., Canada (French: ) is a country in the northern half of North America. Its ten provinces and three territories extend from the Atlantic to the Pacific and northward into the Arctic Ocean, covering , making it the world's second-largest country by total area and the fourth-largest country by land area. Canada's border with the United States is the world's longest land border. The majority of the country has a cold or severely cold winter climate, but southerly areas are warm in summer. Canada is sparsely populated, the majority of its land territory being dominated by forest and tundra and the Rocky Mountains. About four-fifths of the country's population of 36 million people is urbanized and live near the southern border. Its capital is Ottawa, its largest city is Toronto; other major urban areas include Montreal, Vancouver, Calgary, Edmonton, Quebec City, Winnipeg and Hamilton., British Columbia (BC) is the westernmost province of Canada, with a population of more than four million people located between the Pacific Ocean and the Rocky Mountains.
British Columbia is also a component of the Pacific Northwest and the Cascadia bioregion, along with the U.S. states of Idaho, Oregon, Washington and Alaska., The "Champs International Skating Centre of British Columbia" (formerly known as the 'BC Centre of Excellence') is one of two major figure skating training centers in Canada. Located in Burnaby, British Columbia, it is home to many great national and international skaters. The programs there are overseen by a staff, including Joanne McLeod, who coaches 3-time Canadian men's national champion Emanuel Sandhu; Bruno Marcotte, who competed at the 2002 Winter Olympics; Victor Kraatz, the 2003 World Champion in ice dancing, and Maikki Uotila, who was a national champion in Finland. The center operates out of Canlan Ice Sports Burnaby 8 Rinks. Notable skaters who train there include Emanuel Sandhu, Mira Leung, Allie Hann-McCurdy & Michael Coreno, Jessica Millar & Ian Moram, Jeremy Ten, and Kevin Reynolds. This skating school is sometimes known as a training site for international competitors to practice for competitions in Vancouver. Champs International hosts its annual competition known as the BC/YK SummerSkate Competition every August., Shae-Lynn Bourne, MSC (born January 24, 1976) is a Canadian ice dancer. In 2003, she and partner Victor Kraatz became the first North American ice dancers to win a World Championship. They competed at three Winter Olympic Games, placing 10th at the 1994 Winter Olympics, 4th at the 1998 Winter Olympics, and 4th at the 2002 Winter Olympics., Vancouver, officially the City of Vancouver, is a coastal seaport city on the mainland of British Columbia, Canada, and the most populous city in the province., Subject: maikki uotila, Relation: country_of_citizenship, Options: (A) american (B) british (C) canada (D) finland (E) montreal
Solution: finland
Explanation: This is a good example, as maikki uotila is citizen of the finland.
Problem: Context: Mothra is one of Tohos most popular monsters, and second only to Godzilla in its total number of film appearances. Polls taken during the early 1990s indicated that Mothra was particularly popular among women, who were at the time the largest demographic among Japan's movie-going audience, a fact that prompted the filming of 1992's "Godzilla vs. Mothra", which was the best-attended Toho film since "King Kong vs. Godzilla". IGN listed Mothra as #3 on their "Top 10 Japanese Movie Monsters" list, while "Complex" listed the character as #7 on its "The 15 Most Badass Kaiju Monsters of All Time" list., The Cold War was a state of political and military tension after World War II between powers in the Eastern Bloc (the Soviet Union and its satellite states) and powers in the Western Bloc (the United States, its NATO allies and others). Historians do not fully agree on the dates, but a common timeframe is the period (the second half of the 20th century) between 1947, the year the Truman Doctrine (a U.S. policy pledging to aid nations threatened by Soviet expansionism) was announced, and 1991, the year the Soviet Union collapsed., Concept and creation.
Mechagodzilla was conceived in 1974 as a more serious villain than its immediate two predecessors, Gigan and Megalon, whose films were considered creative disasters. According to Tomoyuki Tanaka, Mechagodzilla was inspired by both Mechani-Kong from the previous Toho film "King Kong Escapes" and the robot anime genre, which was popular at the time. Effects director Teruyoshi Nakano also felt that a mechanical monster was cheaper to construct than the mutated animals Godzilla had previously faced. As the resulting "Godzilla vs. Mechagodzilla" proved to be a greater critical success than previous 1970s Godzilla films, the character was revived in 1975's "Terror of Mechagodzilla". The film's screenplay was based on the winning entry of a story-writing competition won by Yukiko Takayama, who continued the darker tone of the previous film by adding the subplot of Mechagodzilla being cybernetically connected to a young woman. Mechagodzilla's design remained largely unchanged from its previous appearance, though it was made to look thinner and more angular, with a darker sheen and an MG2 insignia emblazoned on its upper arms. The film's original draft was going to have Mechagodzilla destroy Tokyo utterly, though the destruction was cut down for budgetary reasons., Toho's most famous creation is Godzilla, who features in 28 of the company's films. Godzilla, Mothra, King Ghidorah, Mechagodzilla, and Rodan are described as Toho's Big Five because of the monsters' numerous appearances in all three eras of the franchise, as well as spin-offs. Toho has also been involved in the production of numerous anime titles. Its subdivisions are Toho-Towa Distribution, Toho Pictures Incorporated, Toho International Company Limited, Toho E. B. Company Limited, and Toho Music Corporation & Toho Costume Company Limited. The company is the largest shareholder (7.96%) of Fuji Media Holdings Inc., The earliest commercial Japanese animation dates to 1917, and production of anime works in Japan has since continued to increase steadily. The characteristic anime art style emerged in the 1960s with the works of Osamu Tezuka and spread internationally in the late twentieth century, developing a large domestic and international audience. Anime is distributed theatrically, by television broadcasts, directly to home media, and over the Internet. It is classified into numerous genres targeting diverse broad and niche audiences., With the nuclear bombings of Hiroshima and Nagasaki and the "Lucky Dragon 5" incident still fresh in the Japanese consciousness, Godzilla was conceived as a metaphor for nuclear weapons. As the film series expanded, some stories took on less serious undertones portraying Godzilla as an antihero while other plots still portrayed Godzilla as a destructive monster; sometimes the lesser of two threats who plays the defender by default but is still a danger to humanity. With the end of the Cold War, several post-1984 Godzilla films shifted the character's portrayal as a symbol of nuclear weapons to that of modern Japan's forgetfulness over its imperial past, natural disasters and the overall human condition., The Return of Godzilla , released in Japan as Godzilla ( Gojira ) , is a Japanese science fiction kaiju directed by Koji Hashimoto and a reboot of Toho 's Godzilla franchise . It is the 16th film in the franchise and the first film in the Heisei series , despite the film having been produced during the Shwa period . The film serves as direct sequel to the original 1954 Godzilla film , completely ignoring the events of the Showa series , which saw the franchise return to the darker tone and themes of the 1954 film and returned Godzilla to his destructive antagonistic roots . In 1985 , New World Pictures released Godzilla 1985 , a heavily re-edited American adaptation of the film which included additional footage , which featured Raymond Burr reprising his role from Godzilla , King of the Monsters ! , which was also produced , filmed , and edited with the same techniques ., Subject: the return of godzilla, Relation: publication_date, Options: (A) 1917 (B) 1947 (C) 1974 (D) 1975 (E) 1984 (F) 1991 (G) 1992 (H) 28 (I) 3 (J) 5
Solution: | 1984 | 4 | NIv2 | task1296_wiki_hop_question_answering | fs_opt | [
0.37707579135894775,
0.635759711265564,
-0.7493605017662048,
-0.3068743348121643,
0.43110954761505127,
0.21439018845558167,
0.7458597421646118,
0.806928277015686,
-0.30264005064964294,
0.09115409851074219,
-0.09257832169532776,
0.6597260236740112,
-0.6504659652709961,
0.5093403458595276,
... |
In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Q: Context Word: wool.
Question: Wool was used to make sweaters but not pants, as the _ were a good garment for this material.
Answer: sweaters.
A: | Wool was used to make sweaters but not pants, as the _ were a poor garment for this material. | 4 | NIv2 | task034_winogrande_question_modification_object | zs_opt | [
0.6492623090744019,
0.2177061140537262,
-0.029669329524040222,
-0.012406640686094761,
0.5154609680175781,
-0.6840716600418091,
0.8208329677581787,
1.0390686988830566,
-0.9067996740341187,
-0.2648307681083679,
-0.6501305103302002,
0.18775513768196106,
-0.5410580635070801,
0.1227465718984603... |
Given a sequence of actions to navigate an agent in its environment, provide the correct command in a limited form of natural language that matches the sequence of actions when executed. Commands are lowercase and encapsulate the logic of the sequence of actions. Actions are individual steps that serve as the building blocks for a command. There are only six actions: 'I_LOOK', 'I_WALK', 'I_RUN', 'I_JUMP', 'I_TURN_LEFT', and 'I_TURN_RIGHT'. These actions respectively align with the commands 'look', 'walk', 'run', 'jump', 'turn left', and 'turn right'. For commands, 'left' and 'right' are used to denote the direction of an action. opposite turns the agent backward in the specified direction. The word 'around' makes the agent execute an action while turning around in the specified direction. The word 'and' means to execute the next scope of the command following the previous scope of the command. The word 'after' signifies to execute the previous scope of the command following the next scope of the command. The words 'twice' and 'thrice' trigger repetition of a command that they scope over two times or three times, respectively. Actions and commands do not have quotations in the input and output.
Input: Consider Input: I_TURN_RIGHT I_RUN I_TURN_RIGHT I_RUN I_TURN_RIGHT I_WALK
Output: walk right after run right twice
Input: Consider Input: I_JUMP I_JUMP I_RUN I_RUN I_RUN
Output: jump twice and run thrice
Input: Consider Input: I_TURN_RIGHT I_WALK I_TURN_RIGHT I_WALK I_TURN_LEFT I_TURN_LEFT I_TURN_LEFT I_TURN_LEFT I_TURN_LEFT I_TURN_LEFT
| Output: turn opposite left thrice after walk right twice
| 2 | NIv2 | task129_scan_long_text_generation_action_command_short | fs_opt | [
0.3440473973751068,
1.0036789178848267,
-0.5712660551071167,
0.3169982433319092,
-0.14225003123283386,
0.17965570092201233,
-0.03556324541568756,
0.619566798210144,
-0.2963279187679291,
-0.11085119098424911,
-0.8737444877624512,
-0.38860881328582764,
-0.5808298587799072,
-0.114670217037200... |
You are supposed to identify the category of a high-school level math question. There are five possible categories (1) algebra (2) arithmetic (3) measurement (4) numbers, and (5) probability. Use the following guidelines: (1) 'algebra' questions will typically contain letter variables and will ask you to find the value of a variable (2) 'arithmetic' questions will ask the sum, difference, multiplication, division, power, square root or value of expressions involving brackets (3) 'measurement' questions are questions that ask to convert a quantity from some unit to some other unit (4) 'numbers' questions will be about bases, remainders, divisors, GCD, LCM etc. (5) 'probability' questions will ask about the probability of the occurrence of something. A question must belong to only one category.
Q: Suppose 17 = -v + 2*f + 1, 0 = 5*v - 3*f + 52. Let i be ((-12)/v + -2)*-4. What is the greatest common divisor of 4 and i?
A: | numbers | 4 | NIv2 | task834_mathdataset_classification | zs_opt | [
0.07215319573879242,
0.6330262422561646,
-0.200929656624794,
-0.12907658517360687,
0.08868415653705597,
0.22102606296539307,
0.42262476682662964,
0.4218430817127228,
0.6856926679611206,
-0.6673592925071716,
-1.1566071510314941,
0.2492876946926117,
-0.5703511834144592,
0.29978233575820923,
... |
Detailed Instructions: In this task, you are given one English sentence. The major part of the sentences talk about health-related topics, but some of the them describe an organization and its activities. Your job is to translate the given sentences into Swedish.
Problem:Conversely, the death rate of acute infection can reach 2% in the elderly.
Solution: | Däremot kan dödligheten vid akut infektion nå 2 procent hos äldre. | 8 | NIv2 | task1395_europa_ecdc_tm_en_sv_translation | zs_opt | [
-0.24401181936264038,
0.8729255199432373,
-0.05340481549501419,
-0.35449567437171936,
-0.2700972557067871,
-0.4501822292804718,
0.2655525803565979,
0.9675561189651489,
0.18628115952014923,
0.5550363063812256,
-0.48665082454681396,
0.7675957679748535,
-0.3065437078475952,
0.3238599598407745... |
Definition: In this task, you're given a sentence and your task is to generate a simplified sentence that focuses on lexical paraphrasing from the original sentence given in the input.
Input: mercury is similar in appearance to the moon : it is heavily cratered with regions of smooth plains , has no natural satellites and no substantial atmosphere.
Output: | mercury , which looks like the moon with its many craters and areas of smoothness , has no orbiting bodies around it and a minimal atmosphere. | 2 | NIv2 | task934_turk_simplification | zs_opt | [
-0.6271986961364746,
1.2718439102172852,
-0.3472362160682678,
-1.3902698755264282,
-0.24854080379009247,
-0.6085841655731201,
0.4168984591960907,
0.7129578590393066,
0.5441217422485352,
-0.2435617446899414,
-0.689061164855957,
-0.38607150316238403,
-0.17147479951381683,
0.410398006439209,
... |
In this task, you are given the abstract of a research paper. Your task is to generate a summary of this abstract. Your summary should not be very short, but it's better if it's not more than 30 words.
Within many machine learning algorithms, a fundamental problem concerns efficient calculation of an unbiased gradient wrt parameters $\boldsymbol{\gamma}$ for expectation-based objectives $\mathbb{E}_{q_{\boldsymbol{\gamma}} (\boldsymbol{y})} [f (\boldsymbol{y}) ]$. Most existing methods either ($i$) suffer from high variance, seeking help from (often) complicated variance-reduction techniques; or ($ii$) they only apply to reparameterizable continuous random variables and employ a reparameterization trick. To address these limitations, we propose a General and One-sample (GO) gradient that ($i$) applies to many distributions associated with non-reparameterizable continuous {\em or} discrete random variables, and ($ii$) has the same low-variance as the reparameterization trick. We find that the GO gradient often works well in practice based on only one Monte Carlo sample (although one can of course use more samples if desired). Alongside the GO gradient, we develop a means of propagating the chain rule through distributions, yielding statistical back-propagation, coupling neural networks to common random variables. | a Rep-like gradient for non-reparameterizable continuous/discrete distributions; further generalized to deep probabilistic models, yielding statistical back-propagation | 0 | NIv2 | task668_extreme_abstract_summarization | zs_opt | [
-0.779327392578125,
0.46533405780792236,
-0.3425726890563965,
-0.7909670472145081,
0.1832834780216217,
-0.6542367339134216,
-0.08579117059707642,
1.2703218460083008,
0.23632675409317017,
-0.19849523901939392,
-1.4193904399871826,
0.21469901502132416,
-0.26331859827041626,
-0.05157693848013... |
Detailed Instructions: In this task, you are given a public comment from online platforms. You are expected to classify the comment into two classes: insult and non-insult. Insult is any lanugage or act that is disrespectful or scornfully abusive.
Problem:Comment: America is America!
-
Canada is... whatever this place is.
Solution: | Non-insult | 8 | NIv2 | task328_jigsaw_classification_insult | zs_opt | [
0.00846291333436966,
0.5272957682609558,
0.6873526573181152,
-0.2805217206478119,
-0.2506423592567444,
0.04747255519032478,
0.951164722442627,
1.0047247409820557,
0.27496767044067383,
0.18055397272109985,
0.18331347405910492,
-0.2415410578250885,
-0.16728782653808594,
-0.6558690071105957,
... |
Part 1. Definition
Given an trivia question precisely answer the question with a word/phrase/name. External resources such as Wikipedia could be used to obtain the facts.
Part 2. Example
Which writer rode Devon Loch in the 1956 Grand National?
Answer: dick francis
Explanation: The answer is precise and accurately answers the question based on a fact. This is a good example
Part 3. Exercise
Who invented Meccano in 1900?
Answer: | frank hornby | 7 | NIv2 | task898_freebase_qa_answer_generation | fs_opt | [
0.33136770129203796,
0.25026166439056396,
-0.23962877690792084,
-0.19272278249263763,
-0.5944379568099976,
-0.24136686325073242,
0.9378604888916016,
-0.8804565072059631,
0.3022505044937134,
-0.8722628951072693,
-1.3355004787445068,
0.7036432027816772,
-0.6158618927001953,
0.347571790218353... |
Given a sentence in Vietnamese, generate a new Vietnamese sentence by performing small changes on the sentence. Here, make sure that the changes are semantically related and syntactically similar to the input. And the generated sentence should have high commonsense plausibility, that is to have reasonable probability of it being true.
Let me give you an example: Anh đang muốn lấy một cái mũ trong một cái hộp.
The answer to this example can be: Anh không sao cả khi nhìn thấy một cái mũ trong một cái hộp.
Here is why: This is a good change in the input. The created sentence is semantically similar to the input as both are talking about a hat in a box and the changes in the sentence follows the commonsense knowledge.
OK. solve this:
Sự ảnh hưởng của việc cắn con chó là con chó đi đến nơi trú ẩn.
Answer: | Ảnh hưởng của việc dắt chó đi dạo là con chó đi vào phòng tắm. | 8 | NIv2 | task411_mickey_vi_sentence_perturbation_generation | fs_opt | [
0.18443085253238678,
0.14769892394542694,
-0.12022099643945694,
0.08673413842916489,
-0.18448340892791748,
-1.4011943340301514,
0.21108412742614746,
1.4763195514678955,
-0.08195166289806366,
0.10167434811592102,
-0.6554951667785645,
0.24002136290073395,
-0.5603336095809937,
-0.239467203617... |
In this task you are given an anecdote. You must find out whether the anecdote is something the author has done or is considering doing. Label the instances as "HISTORICAL" when the author has already done something and label them as "HYPOTHETICAL" when the author is considering doing something in the future.
Example Input: I received a save the date for a wedding. It was from a friend that lives across the country and possibly didn't expect that I would come. However, we do visit that city every summer (my husband's family is from there) and because I got the save the date, I planned the trip around the wedding date so we could attend. Today I got the official invitation and only I am invited, but not my spouse of 5 years. She has met my spouse many times but they are not close. They get along fine though. I am guessing that she doesn't have enough space or doesn't want to pay for an additional meal, especially if she wasn't expecting me to come. However, I know the venue minimum is 100 guests so it's not a very small wedding. I have a couple friends going who's SOs are invited, and I have not heard of anyone else's spouse being excluded.
&#x200B;
WIBTA if I clarified with the bride that my husband is not invited? It could possibly be oversight but he was not in the online system. It could have been me messing up the save the date but I think all I had to do was enter my mailing address so I'm not sure that's possible. I'm not sure if it would be rude to inquire.
&#x200B;
WIBTA if I didn't attend if he's not invited? WIBTA to my husband if I did attend? I will have friends there and although they will have their SOs, I'm sure I would still have a good time if I went. I have heard that it's rude to attend a wedding that your spouse is not invited to, although my husband would be slightly sad but support me going.
Example Output: HYPOTHETICAL
Example Input: My girlfriend and I live in Denver with another roommate in a duplex, outside of downtown and in a residential area with a lot of street parking. Her parents don't live in the continental US, but are currently here vacationing. They want to leave their Jeep with us from now until April/May, and just leave it parked on our street.
&#x200B;
I don't want to have the Jeep just parked in front of our place for 4-5 months, and would rather have them rent a place to store it until they come back in the spring. Admittedly, on our block there is always parking space, so we wouldn't be fucking our neighbors over and taking up a valuable parking space...but her dad does this all the time. He has two motorcycles in the states--one in a friend's garage and another here in Denver in storage that we periodically check on, plus an RV here as well that we do the same for.
&#x200B;
I'd be OK with this if it was for a few weeks or a month, but I don't want it here for half the year. That is half a year that we have to move it for street cleaning, snow plowing, etc. Additionally, we are going to be moving to a new place in January...we don't know where it will be because we are still looking, it will likely be in the same neighborhood with the same parking situation but some of the row homes we are looking at have limits on the number of vehicles per household. He just has shit spread out all over the US with friends and family taking care of all of it, and I feel like he takes advantage of all of us.
Example Output: HISTORICAL
Example Input: The whole #thotaudit fiasco has me hoping that sex workers who don't pay taxes on their earnings deserve punishments. According to Twitter that makes me an asshole incel virgin woman hater. (seriously). I definitely don't like tax but that doesn't mean that they don't have to pay it because they are female. (Obviously there are some male ones and they should pay too but ultimately everyone defends them as their "fellow women") So AITA? (optional WIBTA: for reporting them to the IRS?)
Example Output: | HISTORICAL
| 3 | NIv2 | task282_scruples_event_time | fs_opt | [
-0.01702222414314747,
0.5456341505050659,
0.02204146236181259,
0.511029839515686,
0.5900551080703735,
-0.31627169251441956,
0.9006801843643188,
0.9368590116500854,
-0.21743281185626984,
-0.14548936486244202,
-0.3879217207431793,
0.9697943925857544,
-0.3014380633831024,
0.26956039667129517,... |
The task is about reading the given story and question, then finding an answer to the given question. Based on the passage provided and the given question, you should identify the shortest continuous text span from the passage that serves as an answer to the given question. Avoid answers that are incorrect or provides incomplete justification for the question.
Q: story: The Crimean War was a military conflict fought between October 1853 – March 1856 in which Russia lost to an alliance of France, the United Kingdom, the Ottoman Empire, and Sardinia. The immediate cause involved the rights of Christian minorities in the Holy Land, which was controlled by the Ottoman Empire. The French promoted the rights of Catholics, while Russia promoted those of the Eastern Orthodox Christians. The longer-term causes involved the decline of the Ottoman Empire and the unwillingness of the United Kingdom and France to allow Russia to gain territory and power at Ottoman expense. It has widely been noted that the causes, in one case involving an argument over a key, have never revealed a "greater confusion of purpose", yet led to a war noted for its "notoriously incompetent international butchery."
While the churches eventually worked out their differences and came to an initial agreement, both Nicholas I of Russia and Napoleon III refused to back down. Nicholas issued an ultimatum that the Orthodox subjects of the Empire be placed under his protection. Britain attempted to mediate, and arranged a compromise that Nicholas agreed to. When the Ottomans demanded changes, Nicholas refused and prepared for war. Having obtained promises of support from France and Britain, the Ottomans officially declared war on Russia in October 1853., question: What was Britain's response?
A: | Britain attempted to mediate | 4 | NIv2 | task918_coqa_answer_generation | zs_opt | [
1.0494037866592407,
0.3488912880420685,
-0.9352480173110962,
-0.7357572317123413,
0.19582286477088928,
-0.6279568672180176,
0.662574052810669,
0.5801277160644531,
0.269262433052063,
0.6206438541412354,
-0.21136683225631714,
-0.06848413497209549,
-0.6540422439575195,
-0.19243109226226807,
... |
Part 1. Definition
In this task, you will be presented with a question having multiple possible answers in Portuguese language. And you should choose a most suitable option out of "A", "B", "C", "D", and "E" based on your commonsense knowledge.
Part 2. Example
Question: Foi um presente popular e comum este ano, por isso dirigiu-se para o que seria fácil de encontrar?
Options: (A) árvore de natal (B) meia de natal (C) loja especializada (D) armário (E) grande armazém
Answer: E
Explanation: The predicted answer is correct because out of given many options only department stores are easy to find.
Part 3. Exercise
Question: Onde é que uma bilheteira estaria a vender bilhetes para algo que está a acontecer lá fora?
Options: A teatro de cinema B distrito teatral C teatro D ballpark E rua
Answer: | D | 7 | NIv2 | task1137_xcsr_pt_commonsense_mc_classification | fs_opt | [
0.4167851209640503,
0.5400398969650269,
-0.36093541979789734,
0.035767748951911926,
-0.09050551801919937,
-0.09941253066062927,
-0.03390713036060333,
1.0753707885742188,
0.4591938257217407,
0.11678662896156311,
-0.2642465829849243,
0.3346869945526123,
-0.07261200249195099,
-0.2006994187831... |
Q: Given a sentence in the Indonesian(Bahasa variant), provide an equivalent translation in Japanese that retains the same meaning through the translation. In translation, keep numbers as it is.
Pelayanan microblogging Twitter telah diganti dengan dukungan perangkat lunak untuk sensor pemberitahuan.
A: | マイクロブログサービスのツイッターは投稿の検閲のサポートソフトに変更した。 | 7 | NIv2 | task1116_alt_id_ja_translation | zs_opt | [
-1.216434359550476,
-0.11212952435016632,
0.12074518203735352,
-0.36964231729507446,
0.09349039196968079,
-0.9504573345184326,
0.4217028021812439,
0.6418560743331909,
0.19933566451072693,
0.11691605299711227,
-0.2011757791042328,
0.056193046271800995,
-0.517217755317688,
0.5416348576545715... |
Q: You are asked to create a question containing a blank (_), based on the given context word. Your question must contain two related but different objects; for example "trophy" and "suitcase". The expected answer to your question must be one of the objects present in the sentence. The expected answer must not be associated with any specific word in the question; instead it should depend on the context present in the question. The expected answer should not be equally likely to fill the blank. For your question, there should be a agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. Do not use animals or proper nouns (e.g., New York, Macbook, Jeff Bezos, McDonald's, ...) as your objects. Avoid repeating the same style, pattern or phrases in each question, try to increase diversity by varying sentence structure, blank placement etc. Your question must contain at least 15 and at most 30 words. You must utilize the given context word while writing the question. Your question must contain only one blank. Make sure that Object X and Y have the same number e.g. when ObjectX is singular, Object Y must be singular, too. The two objects (Object X & Object Y) should be used ONCE in your question. Here is a list of attributes and associated pair of contrastive words which may be used to create a valid question using the objects. You can use either of the contrastive words, but not both. You should think about more such attributes and associated words and use them in your question.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Context Word: crash.
A: | The car crashed into a tree but only the tree was damaged, because the _ is made from stronger material. | 7 | NIv2 | task031_winogrande_question_generation_object | zs_opt | [
0.7745749950408936,
0.43330809473991394,
0.37541553378105164,
0.021555930376052856,
0.7008178234100342,
-0.8267076015472412,
0.6513075828552246,
1.3720802068710327,
-0.2299671471118927,
-0.44896435737609863,
-0.33582550287246704,
0.40463513135910034,
-0.25085756182670593,
0.042744811624288... |
Q: In this task, you're given the title of a story consisting of five sentences, numbered 1 through 5. Your job is to determine which two sentences need to be swapped sentences in order to make a story that makes complete sense and is befittingly titled. Indicate your answer using the numbers of the two sentences in order, such as '34' or '25'. The first digit refers to the sentence which should come first in the story.
Title: Movies with girlfriend. Sentence 1: My girlfriend and I wanted to do something yesterday. Sentence 2: It was a good movie. Sentence 3: We decided to go to the movies. Sentence 4: I was glad we went. Sentence 5: We spent hours thinking until she had an idea.
A: | 25 | 7 | NIv2 | task218_rocstories_swap_order_answer_generation | zs_opt | [
-0.6719970107078552,
0.29370006918907166,
-0.04029276221990585,
-0.4403529763221741,
-0.2145521193742752,
-0.09647130966186523,
0.7048813104629517,
0.5077865123748779,
0.17793476581573486,
-0.42356061935424805,
-0.6213420629501343,
-0.7904806137084961,
-0.9882987141609192,
-0.0414385721087... |
Given the task definition, example input & output, solve the new input case.
In this task, you are given a public comment from online platforms. You are expected to classify the comment into two classes: sexual-explicit and non-sexual-explicit. A comment is considered sexual-explicit if it explicitly portrays sexual matters.
Example: Comment: President lied under oath about consensual sex...
Not the same thing as rape.
Output: Sexual-explicit
This comment explicitly mentions "consensual sex" and "rape", and thus is sexual-explicit.
New input case for you: Comment: I just read Harvard Hosts Anal Sex Workshop Entitled ‘What What in the Butt’, followed for a queer only talent show at Stevenson's.
Actually that's the democrat's dream, one washroom for all, same gender sex, self gratifying anal adventures. Just look at Hollywood, the ingrates get into consensual sex, then accuse their one time partner with a sexual offense or a big pay out, your choice.
Universities and Colleges had become an extension of decadent Hollywood celebrities.
Output: | Sexual-explicit | 1 | NIv2 | task323_jigsaw_classification_sexually_explicit | fs_opt | [
-0.455738365650177,
0.03862965852022171,
-0.07322835177183151,
0.2948594093322754,
-0.022459181025624275,
-0.0005546582397073507,
-0.49185508489608765,
0.6569234728813171,
-0.03469347208738327,
0.07614396512508392,
-0.06242445111274719,
0.4037103056907654,
-0.3188358545303345,
-0.980777442... |
In this task, you are given an input list. A list contains several comma-separated items written within brackets. You need to return the position of all the alphabetical elements in the given list in order. Assume the position of the 1st element to be 1. Return -1 if no alphabetical element is in the list.
Q: ['8307', '393', 'T', '9117', 'l', '2541', 'F', 'F', 'U', 'T', '5991', '8377', 'm']
A: 3, 5, 7, 8, 9, 10, 13
****
Q: ['9963', '3165', 'K', '7637', 'A', '4361', '3759', '7485', 'r', 'S', '5471', 'A', '3153', 'z', 'H', '5903', '7783', '4333', 'e', '2003', '2719', 'B', 'v', 'h', 'T', '3453', 'R', '8459', '9619', '9603']
A: 3, 5, 9, 10, 12, 14, 15, 19, 22, 23, 24, 25, 27
****
Q: ['7777', 'e', '3141', 'R']
A: | 2, 4
****
| 4 | NIv2 | task506_position_of_all_alphabetical_elements_in_list | fs_opt | [
-0.2801596224308014,
0.051068224012851715,
-0.20882710814476013,
-0.13950788974761963,
0.18400385975837708,
-0.5789771676063538,
0.6071479320526123,
0.5987780094146729,
-0.9408360719680786,
0.6017836332321167,
-0.5717459321022034,
0.04858081787824631,
-0.13380438089370728,
0.18255612254142... |
Instructions: Two analogies on manipulating items in a kitchen is given in the form "A : B. C : ?". The phrase "A : B" relates item A to its appropriate manipulation B. Your task is to replace the question mark (?) with the appropriate manipulation of item C, following the "A : B" relation. Your answer should be a verb which shows a way to manipulate the given object.
Input: box : open. peas : ?
Output: | shell | 3 | NIv2 | task1158_bard_analogical_reasoning_manipulating_items | zs_opt | [
0.05405092239379883,
1.1863160133361816,
-1.0109491348266602,
0.7702783346176147,
-0.3421535789966583,
-0.343497097492218,
1.0428001880645752,
0.6161656379699707,
-0.02160121127963066,
-0.2782268822193146,
-0.29042932391166687,
0.21622750163078308,
0.377110093832016,
0.11516457051038742,
... |
Given the task definition and input, reply with output. In this task, you are given an input text which is in one of the 11 languages listed as follows: ['Oriya' 'Malayalam' 'Bengali' 'Gujarati' 'Hindi' 'English' 'Marathi'
'Panjabi' 'Tamil' 'Telugu' 'Urdu']. Identify the language of input text. The input text can be in one of the given languages only. Input sentences must only have one language at a time.
யூ யங் மின், கொரிய குடியரசின் அறிவியல் மற்றும் தகவல் தொடர்பு தொழில்நுட்பத் துறை அமைச்சர்
| Tamil | 5 | NIv2 | task976_pib_indian_language_identification | zs_opt | [
-0.8186547756195068,
0.8527641296386719,
0.3872912526130676,
-0.8115551471710205,
0.3881106972694397,
-0.4250885546207428,
-0.12331300973892212,
-0.24051767587661743,
-0.0558912456035614,
-0.39391472935676575,
0.25136011838912964,
0.40997809171676636,
0.36746707558631897,
0.354481041431427... |
You are given an open-domain question from an open movie database. Your task is to provide an answer to that question. Try to only include the answer. Do not put it in a sentence.
[Q]: what is the genre for the film Overlord?
[A]: Drama, War, History
[Q]: who are the actors in Dog Nail Clipper?
[A]: Peter Franzén, Ville Virtanen, Taisto Reimaluoto, Ahti Kuoppala
[Q]: what sort of film is Jewel Robbery?
[A]: | Comedy, Crime, Romance
| 5 | NIv2 | task615_moviesqa_answer_generation | fs_opt | [
-0.5283021926879883,
0.3761868178844452,
-0.6377323865890503,
0.20502206683158875,
-0.4279884099960327,
0.011130076833069324,
0.43761032819747925,
0.5149458646774292,
-0.21339745819568634,
-0.15290866792201996,
0.06909757107496262,
0.2421334981918335,
-1.0106070041656494,
-0.73309254646301... |
Teacher:You are shown a conversation between a user and system. Identify who has spoken the indicated sentence based on the conversation.
Teacher: Now, understand the problem? Solve this instance: Sentence1: Have a nice day! Goodbye. Sentence2: Royal Standard is a Gastropub in East. Is this OK? Sentence3: May I recommend the Grafton Hotel Restaurant, then? They're very British, very expensive, and located in the east part of town. Sentence4: Thank you, goodbye. Sentence5: Yes there are four other options. What kind of food would you like? Sentence6: That sounds like something I would enjoy. Could you give me the address and phone number? Sentence7: It doesn't matter, just not Gastropub. Sentence8: is there any other expensive restaurant in east? Sentence9: Hi, I want an expensive restaurant. East part of town please. Sentence10: Certainly. The Grafton Hotel restaurant is at 619 Newmarket Road Fen Ditton and the number there is 01223 241387. Can I be of further assistance? Question: Who has spoken the sentence 5? (A) User (B) System
Student: | Answer: (B) System | 6 | NIv2 | task638_multi_woz_classification | zs_opt | [
0.22589349746704102,
0.03344464302062988,
0.031169775873422623,
-0.5513172149658203,
-0.023962512612342834,
-0.9550246000289917,
0.33216017484664917,
0.9035073518753052,
-0.16238878667354584,
-0.18236032128334045,
-0.16880999505519867,
-0.44083884358406067,
-0.3930964469909668,
-0.36879819... |
Detailed Instructions: You are given a sentence in Polish. Your job is to translate the Polish sentence into Portugese.
Q: Między innymi, jeśli chcesz się pochwalić swoim bogactwem, wpływami, zawsze lepiej posiadać oryginał niż podróbkę, ponieważ zawsze będzie mniej oryginałów niż podróbek.
A: | Entre outras coisas, se queres exibir o quão rico és, o quão poderoso és, é sempre melhor adquirir um original do que uma falsificação porque sempre existirão menos originais do que falsificações. | 9 | NIv2 | task1264_ted_translation_pl_pt | zs_opt | [
-0.7964826822280884,
1.190704584121704,
0.17737361788749695,
-0.1492769420146942,
-0.5602430105209351,
-0.9524027109146118,
0.17096060514450073,
0.678267776966095,
0.5717795491218567,
-0.2137538492679596,
-0.3437894284725189,
0.74639892578125,
-0.744825541973114,
0.0060580335557460785,
0... |
Detailed Instructions: In this task, you are given two strings A,B. Find the longer of the two lists, convert it to lowercase, and return all the unique alphabets used in it. The two input strings are never equal.
Problem:wgyNiJhZRplmTwGkvQbDOG, iipZCigqRplmTwGkSFLcz
Solution: | b, d, g, h, i, j, k, l, m, n, o, p, q, r, t, v, w, y, z | 8 | NIv2 | task756_find_longert_substring_and_return_all_unique_alphabets_in_it | zs_opt | [
0.18222978711128235,
1.181627631187439,
0.17196789383888245,
-0.5095256567001343,
0.07103102654218674,
-0.38025569915771484,
1.5754187107086182,
-0.23150157928466797,
-0.15924619138240814,
-0.547792911529541,
-0.604631781578064,
-0.1241503655910492,
-0.12908078730106354,
-0.143714725971221... |
Teacher: Given a post that is a real-life anecdote of a complex ethical situation and a question asks if AUTHOR, NOBODY, EVERYBODY, or OTHER is wrong in the situation. Answer the question and classify your answers into yes or no.
Teacher: Now, understand the problem? If you are still confused, see the following example:
Is AUTHOR wrong in the post ? POST : Pretty much was hanging out with two friends. I was watching the Lakers game. They won at the very last minute and I immediately hopped up and celebrated. Before I could even finish celebrating one friend started going on about how the Lakers suck. It just pissed me off so much I screamed at her to get out my room until she left.
She knows I'm a huge fan and she likes them too. She was only saying it to antagonize me. Am I the dick here?
Solution: yes
Reason: It is very rude to shout at someone and ask her to leave the room even if her comments may have antagonized the author. Hence author is at fault here.
Now, solve this instance: Is AUTHOR wrong in the post ? POST : So I live in Belgium and my family is not Belgian.
I go to one of the best high schools in Belgium and I study a difficult direction with very intelligent classmates except they are very very childish for 15 year Olds. Today I was just going to my class when a classmate pushed me and told me that he gets to pass first because he's from Belgium. I ve been getting racist remarks all year for example hey 'op' we want our airport back and we want the towers back.
Yesterday was my birthday and it went bad nobody showed from my family so I was alone with just my parents so I was already mad and this just pushed it. Therefor I don't want to talk to them. And since I'm not from Belgium I'm good at languages like French and better at English than most well we had those classes today and usually I make their assignments (I don't know why) but today I was to triggered to make it. AITA or not?
Student: | no | 2 | NIv2 | task502_scruples_anecdotes_whoiswrong_verification | fs_opt | [
0.22007820010185242,
-0.06659111380577087,
0.39908793568611145,
0.2722562551498413,
0.29526782035827637,
-0.2992026209831238,
0.04252617806196213,
0.7193556427955627,
0.3423924446105957,
-0.13863776624202728,
-0.03866685926914215,
0.26924118399620056,
-0.7073930501937866,
-0.51057308912277... |
instruction:
In this task, you're given a context passage, followed by a question that needs to be answered. Based on the paragraph, you must write unambiguous answers to the questions and your answer must refer to a specific phrase from the paragraph. If multiple answers seem to exist, write the answer that is the most plausible.
question:
Ash arrived at the restaurant to meet his date for dinner.
Question: What will Ash want to do next?
answer:
talk to the host to get a table
question:
Austin stood on the karaoke stage and sang in a high pitch sound.
Question: How would Others feel as a result?
answer:
like turning the music off
question:
One event I hate , is high jump . I've had a bad history with high jump * evil stare * . Anyway , yesterday I went to church with my friend . I slept over with my friend so I had to go to church with her .
Question: What's a possible reason the writer hates high jump ?
answer:
| Because they are not good at it .
| 9 | NIv2 | task595_mocha_answer_generation | fs_opt | [
0.3973597288131714,
0.794426441192627,
-0.21158438920974731,
0.028985977172851562,
-0.45370447635650635,
-0.26877254247665405,
0.47104233503341675,
0.33436042070388794,
0.010321211069822311,
0.0963585376739502,
-0.5285258293151855,
0.03440726920962334,
-0.5268150568008423,
0.29859066009521... |
Instructions: In this task, you're given a fill-in-the-blank question that contains two object names. Additionally, you're given one answer which is one of the objects present in the question. In this task, you need to minimally change the given question so that the answer flips to another object in the question. Remember the question has to be about both objects which are related but different; for example, "trophy" and "suitcase". This task typically involves replacing one word i.e., the 'trigger word' with its antonym (e.g., changing from "small" to "big"). You should not change any content in the given question beyond a word or two, i.e., the trigger word/phrase. The expected answer to your question must not be associated with just the trigger word; instead, it should depend on the context present in the question. The expected answer should not be ambiguous. For your question, there should be an agreed upon answer to fill in the blank. Your generations should NOT contain potentially explicit, offensive, or adult content. In your generated question, retain the object names from the input question. Avoid repeating the same style or phrase in generating your modified question e.g. this task can be always solved using a simple negation i.e. by adding not, never, etc. Instead, try to increase the word diversity. Your question must contain at least 15 and at most 30 words. Your question must have at least 70% overlapping words with the given question. You must utilize the given context word while writing the question. Your question must contain only one blank. The two objects should be used ONCE in your question. Here is a list of attributes and associated contrastive words that may help write contrastive trigger words. Note that contrastive trigger words help in flipping the label.
| Attribute | triggerword | contrastive triggerword |
| age | old | new |
| altitude | low | high |
| area | small | vast |
| brightness | dark | light |
| clarity | obscure | clear |
| cleanness | dirty | clean |
| complexity | simple | complex |
| cost | cheap | expensive |
| density | sparse | dense |
| depth | shallow | deep |
| distance | near | far |
| electric conductivity | low | high |
| flexibility | rigid | flexible |
| granularity | fine | coarse |
| hardness | soft | hard |
| length | short | long |
| magnitude | small | large |
| mass | small | large |
| odor | weak | strong |
| pressure | low | high |
| resistance | low | high |
| shape | round | sharp |
| shape | flat | spiky |
| size | small | large |
| sound | quiet | loud |
| sound pitch | low | high |
| speed | slow | fast |
| stability | unstable | stable |
| strength | weak | strong |
| temperature | low | high |
| texture | smooth | rough |
| thermal conductivity | low | high |
| thickness | thin | thick |
| volume | small | large |
| weight | light | heavy |
| width | narrow | wide |
| location | in | out |
| location | up | down |
| location | above | below |
| location | on | off |
| location | to | from |
Input: Context Word: icy.
Question: While trying to cook Jane set her hot pan into a bowl of icy water. The _ needed its temperature adjusted.
Answer: pan.
Output: | While trying to cook Jane set her hot pan into a bowl of icy water. The _ inadvertently got its temperature adjusted. | 3 | NIv2 | task034_winogrande_question_modification_object | zs_opt | [
0.6907401084899902,
0.17665550112724304,
0.011422928422689438,
-0.10046661645174026,
0.5982875823974609,
-0.6933428049087524,
0.9146716594696045,
1.013892650604248,
-0.794922947883606,
-0.3476032018661499,
-0.355643630027771,
0.17625579237937927,
-0.41913172602653503,
-0.03114432282745838,... |
Definition: You are given a sentence in Polish. Your job is to translate the Polish sentence into Italian.
Input: Chciałbym jej dziś za to podziękować.
Output: | E oggi voglio ringraziarla. | 2 | NIv2 | task1262_ted_translation_pl_it | zs_opt | [
-0.7091370820999146,
0.8040143251419067,
-0.02679845318198204,
0.3270201086997986,
-0.10371017456054688,
-0.7743505239486694,
0.692110002040863,
-0.6008189916610718,
0.6621446013450623,
-0.3345434069633484,
-0.7157613039016724,
0.38673126697540283,
0.2265177071094513,
0.47723984718322754,
... |
instruction:
This is a paraphrasing task. In this task, you're given a sentence and your task is to generate another sentence which express same meaning as the input using different words.
question:
they are n't humans , not anymore . - desert conjurers , cut-throats .
answer:
it 's not people anymore , it 's desert cutthroats .
question:
if-if i do n't continue my treatment , then they 'll , um ... they 'll take him away from me .
answer:
if i do n't continue my treatment , they 'll take me away .
question:
i mean , it might be ... it might be the tie they wear ... or the books they read .
answer:
| maybe by the tie he 's wearing , or the books they read or do n't read .
| 9 | NIv2 | task177_para-nmt_paraphrasing | fs_opt | [
-0.32609808444976807,
0.31699660420417786,
-0.44049957394599915,
-0.5419519543647766,
-0.7683892250061035,
-0.8929078578948975,
0.9191327095031738,
0.489326536655426,
0.33896106481552124,
0.013476971536874771,
-0.05669442564249039,
-0.5753992795944214,
-0.5663184523582458,
0.23374429345130... |
In this task, you're given a pair of sentences, sentence 1 and sentence 2. Your job is to classify the genre of the sentences from the 10 options displayed. Indicate your answer by entering the number of the choice as 1-10.
Q: Sentence 1: Republicans' We're winning the budget fight by convincing Clinton to compromise. Sentence 2: Republicans want to convince Clinton to compromise in the budget fight. Choices: 1. FACE-TO-FACE, 2. GOVERNMENT, 3. LETTERS, 4. 9/11, 5. SLATE, 6. TELEPHONE, 7. TRAVEL, 8. VERBATIM, 9. OUP, 10. FICTION.
A: 4
****
Q: Sentence 1: La Manga has been developed as a resort, with hotels and apartment blocks straddling the 23 km (14 miles) of sand that separate the two seas. Sentence 2: No construction is allowed on the sand between the seas. Choices: 1. FACE-TO-FACE, 2. GOVERNMENT, 3. LETTERS, 4. 9/11, 5. SLATE, 6. TELEPHONE, 7. TRAVEL, 8. VERBATIM, 9. OUP, 10. FICTION.
A: 6
****
Q: Sentence 1: Comfortable modern facilities are offered in unusual black basalt buildings at the Church of Scotland Hospice, which was constructed a century ago as a missionary hospital. Sentence 2: It is modern despite the building being 300 years old. Choices: 1. FACE-TO-FACE, 2. GOVERNMENT, 3. LETTERS, 4. 9/11, 5. SLATE, 6. TELEPHONE, 7. TRAVEL, 8. VERBATIM, 9. OUP, 10. FICTION.
A: | 6
****
| 4 | NIv2 | task198_mnli_domain_classification | fs_opt | [
-0.17098581790924072,
0.2714459002017975,
-0.11819948256015778,
-0.32125943899154663,
-0.05056697875261307,
-0.3111026883125305,
0.12572503089904785,
0.9622898101806641,
-0.15000750124454498,
-0.010416306555271149,
-0.19772599637508392,
0.41563206911087036,
-0.38381898403167725,
-0.4286383... |
Q: In this task, you're given a passage, further information available on a particular linked term from the statement, and a question. Your job is to generate the answer to the question by using the information provided. If there is no clear answer obtainable, output 'none'.
Passage: After Meissner’s abitur at the Humanistisches Gymnasium, in 1910, he began the study of physics and mathematics at University of Tübingen. After three terms, he went to the University of Munich as a student of the experimentalist Wilhelm Röntgen and the theoretician Arnold Sommerfeld. After one year at Munich, he returned to Tübingen to be able to study spectroscopy with Friedrich Paschen. While still a student in 1914, Meissner was able to prove the existence of oxygen lines in the solar spectrum. He was awarded his doctorate in 1915, on a thesis with the title Interferometrische Wellenlängenbestimmung im infraroten Spektralbereich. In 1916 he became an assistant to Edgar Meyer at the University of Zurich, under whom he completed his Habilitation, in 1918, with the Habilitationsschrift title Untersuchungen des Neonspektrums. The following year, he married the Polish physicist Doctor Janka Kohn.
Link Information: Spectroscopy is the study of the interaction between matter and electromagnetic radiation Question: What is the focus of the subject Meissner studied on his return to Tübingen?
A: | Answer: the interaction between matter and electromagnetic radiation | 7 | NIv2 | task238_iirc_answer_from_passage_answer_generation | zs_opt | [
-0.047357745468616486,
0.17654043436050415,
-0.6220595836639404,
-0.22938323020935059,
0.15098316967487335,
0.5897529721260071,
0.42518118023872375,
0.5677455067634583,
0.762898862361908,
0.23571205139160156,
-0.2417018711566925,
0.27465635538101196,
-0.6177811622619629,
-0.098634526133537... |
You are given a sentence in English. Your job is to translate the English sentence into Italian.
His story is their story. | La sua storia è la loro storia. | 0 | NIv2 | task1091_ted_translation_en_it | zs_opt | [
-0.5989606380462646,
0.5874189734458923,
0.023100730031728745,
0.11411227285861969,
-0.39383548498153687,
-0.33953994512557983,
0.431089848279953,
0.3305876851081848,
0.09368453174829483,
-0.21288125216960907,
-0.260878324508667,
0.4490352272987366,
0.13844722509384155,
0.13569003343582153... |
Given the task definition and input, reply with output. This task is about reading the given passage and construct a question about the information present in the passage. Construct a question in such a way that (i) it is unambiguous, (ii) it is answerable from the passage, (iii) its answer is unique (iv) its answer is a continuous text span from the paragraph. Avoid creating questions that (i) can be answered correctly without actually understanding the paragraph and (ii) uses same words or phrases given in the passage.
In 1271, Kublai Khan imposed the name Great Yuan (Chinese: 大元; pinyin: Dà Yuán; Wade–Giles: Ta-Yüan), establishing the Yuan dynasty. "Dà Yuán" (大元) is from the sentence "大哉乾元" (dà zai Qián Yuán / "Great is Qián, the Primal") in the Commentaries on the Classic of Changes (I Ching) section regarding Qián (乾). The counterpart in Mongolian language was Dai Ön Ulus, also rendered as Ikh Yuan Üls or Yekhe Yuan Ulus. In Mongolian, Dai Ön (Great Yuan) is often used in conjunction with the "Yeke Mongghul Ulus" (lit. "Great Mongol State"), resulting in Dai Ön Yeke Mongghul Ulus (Mongolian script: ), meaning "Great Yuan Great Mongol State". The Yuan dynasty is also known as the "Mongol dynasty" or "Mongol Dynasty of China", similar to the names "Manchu dynasty" or "Manchu Dynasty of China" for the Qing dynasty. Furthermore, the Yuan is sometimes known as the "Empire of the Great Khan" or "Khanate of the Great Khan", which particularly appeared on some Yuan maps, since Yuan emperors held the nominal title of Great Khan. Nevertheless, both terms can also refer to the khanate within the Mongol Empire directly ruled by Great Khans before the actual establishment of the Yuan dynasty by Kublai Khan in 1271.
| When did Khan establish the Great Yuan? | 5 | NIv2 | task074_squad1.1_question_generation | zs_opt | [
0.3208916485309601,
0.7519881129264832,
-0.5474201440811157,
-0.7090125679969788,
-0.3498351573944092,
-0.6304408311843872,
-0.026820261031389236,
0.7951682209968567,
0.2512539029121399,
-0.23493345081806183,
-0.49159443378448486,
0.15026170015335083,
-0.7963717579841614,
0.528253197669982... |
This task involves creating questions from a given passage that involve some kind of complex reasoning (including numerical reasoning).
The generated questions must require looking at more than one part of the passage to answer. Try to use a variety of reasoning types in your questions (some of the sample reasoning types are illustrated in the 'positive examples' field) and also have a variety of answer types (spans, numbers, dates). A span is a continuous phrase taken directly from the passage or question. In such questions, the answer spans are recommended to be five words. Questions with answer type "number", are expected to start with "How many". For instance, "How long did an event last?" can be phrased as "How many years did an event last?". To increase diversity, try to create more questions for which the answer will be multiple spans. Please create questions that require AT LEAST 2 arithmetic operations as illustrated in the 'positive examples' field. You are encouraged to try DIFFERENT COMBINATIONS of operations and not just limit to the combinations shown in those examples. The more challenging questions you create, the better it is.
Example input: Passage: The outbreak of the Seven Years' War in Europe in 1756 resulted in renewed conflict between French and British forces in India. The Third Carnatic War spread beyond southern India and into Bengal where British forces captured the French settlement of Chandernagore in 1757. However, the war was decided in the south, where the British successfully defended Madras, and Sir Eyre Coote decisively defeated the French, commanded by Comte de Lally at the Battle of Wandiwash in 1760. After Wandiwash, the French capital of Pondicherry fell to the British in 1761. The war concluded with the signing of the Treaty of Paris in 1763, which returned Chandernagore and Pondichéry to France, and allowed the French to have "factories" in India but forbade French traders from administering them. The French agreed to support British client governments, thus ending French ambitions of an Indian empire and making the British the dominant foreign power in India.
Example output: Which french settlement did the British capture first, Chandernagore or Pondicherry?
Example explanation: This is a good question since it involves a comparison around a particular event, "British capturing the French settlement", which requires looking at more than a single sentence.
Q: Passage: Tampa Bay traveled west to San Francisco for the third time in five seasons. Tampa Bay lost their eighth consecutive meeting at Monster Park, dating to 1980. This game was originally scheduled for 8:15 p.m. on NBC's Sunday Night Football, but on December 10, in accordance with flex-scheduling, the game was moved to the afternoon. Having already clinched a playoff berth, Tampa Bay rested most of their starters in the second half. Despite a late rally, the Buccaneers fell two points short of victory. Tampa Bay controlled most of the first half, capping off their first two drives of the first quarter with field goals by Matt Bryant. On the first play of the second quarter, Shaun Hill connected with Darrell Jackson on a 21-yard touchdown. San Francisco punted on their next drive, and Micheal Spurlock fielded the ball at the 19-yard line. He was tackled and fumbled, while the 49ers recovered. After an instant replay challenge, it was determined that Spurlock was down by contact, and Tampa Bay maintained possession. The drive fizzled though, and ended with a punt. Late in the fourth quarter, Barrett Ruud recovered a 49ers fumble, setting the Buccaneers up on the San Francisco 36-yard line. Four plays later, Jerramy Stevens caught a touchdown pass from Jeff Garcia. Tampa Bay took a 13-7 lead into halftime. On the second-half kickoff, San Francisco recovered a surprise onside kick. The drive came up empty, as they were forced to punt. On Tampa Bay's next possession, however, Luke McCown (who had taken over for the benched Garcia) fumbled and the 49ers recovered at the 14-yard line. In four plays, San Francisco scored a touchdown. Early in the fourth quarter, Nate Clements intercepted a McCown pass, and led to another 49ers touchdown. The 49ers led 21-13, but Tampa Bay still kept hopes alive for a rally. With less than six minutes left in the game, the Buccaneers drove to the 49ers 25-yard line. On 4th down and 6, McCown threw to Michael Clayton for an apparent one-handed touchdown catch, but the ball fell incomplete, and the drive turned over on downs. Tampa Bay managed one final chance to tie the score. Inside the two-minute warning, Tampa Bay drove to the 49ers 24-yard line. McCown rolled out wide to his right, and connected with Stevens for a 24-yard touchdown. McCown then attempted a game-tying two-point conversion. Michael Clayton caught the pass, but one of his feet touched out-of-bounds in the back of the endzone, sealing the game for San Francisco.
A: | How many field goals were scored in the first quarter? | 3 | NIv2 | task026_drop_question_generation | fs_opt | [
1.3176600933074951,
0.20355764031410217,
-0.4857790470123291,
-0.05599774420261383,
0.6723651885986328,
-0.5497642755508423,
0.914991021156311,
0.6398586630821228,
0.33119404315948486,
0.2796242833137512,
-0.22228610515594482,
0.7030824422836304,
-1.085728645324707,
0.21723076701164246,
... |
In this task, you are given a text of many news articles seperated by special token "|||||". Your task is to summarize them.
[Q]: Ronald Dillon, a longtime city Health Department employee, was suspended for 20 days without pay after he repeatedly answered customer service calls in a robot voice, according to the Office of Administrative Trials and Hearings. View Full Caption Thinkstock
NEW YORK CITY — A longtime city Health Department worker was suspended 20 days without pay for answering customer-service calls in a robot voice.
Ronald Dillon, a computer specialist for the agency's IT help desk who assists co-workers and the public with tech-related problems, repeatedly channeled his inner Siri by talking in a "deliberately robotic fashion" when he fielded calls — despite his boss telling him to stop, according to an Office of Administrative Trials and Hearings decision.
Representatives for the Health Department said during a disciplinary hearing before Administrative Law Judge Kara Miller that Dillon talked in the phony voice on at least five occasions between February and April 2013.
Miller's decision says that during the hearing the Health Department played a recording of Dillon speaking to a customer in a "slow, monotone and over-enunciated manner" and saying, "You have reached the Help Desk. This is Mr. Dillon. How may I help you?"
His droid imitation was apparently good enough to fool callers.
One confused customer who spoke to Dillon later called back and told another Health Department worker that she thought "there was a new automated answering system and had hung up when she heard 'the robot' answer the phone because she needed to speak to a human about her issues," the decision says.
The Health Department originally gave Dillon a 34-day suspension for the robot voice, but Miller lowered the penalty to 20 days in her Feb. 14 decision.
Dillon then appealed the decision with the city's Civil Service Commission. Earlier this month the commission upheld Miller's ruling.
During the Feb. 14 hearing, Dillon defended the tone of his voice, claiming that he wasn't mimicking a robot — rather he was just slowly reading from a script his boss instructed him to follow when dealing with customers.
"He contended that he articulates each word because he speaks fast and has a Brooklyn accent, which is sometimes difficult to understand," the judge wrote in her decision.
Dillon also told her that he was not a "people person" and that his boss, Barry Novack, picked on him.
Dillon also told DNAinfo New York on Thursday that he was simply trying to appease his boss by making his voice sound neutral on the phone.
"They objected to the tone of my voice so I made it atonal," he said.
But Miller found that the robot voice was intentional, calling Dillon a "disgruntled" employee who didn't like his supervisor.
"There is a difference between speaking slowly and distinctly and speaking so robotically that callers did not believe that they were speaking to a person," she wrote in her decision.
Dillon has worked for the Health Department since 1976 and has never faced disciplinary action before this case, the decision says.
His job entails responding to employees' computer problems and to tech questions from the public about NYCMED, a web portal that allows healthcare providers to register with the Health Department.
Judge Miller said in her decision that in the past few years, Dillon's job duties changed and Novack became his supervisor.
"He appears to be [a] disgruntled employee who is acting out because of the restructuring that has occurred within the department over the last few years," she wrote.
The judge also found that Dillon didn't follow through on service requests and didn't stick to protocol while handling requests.
Dillon, who in his private life has led grassroots campaigns against a composting center in Spring Creek Park in Brooklyn, said for many years he was a project manager and supervisor in the Health Department.
But he said about three years ago, against his objections, he was ordered to transfer to the help desk job, even though he didn't have the requisite customer service background.
"It's a bullying-in-the-workplace issue," he said.
He said he has asked for a transfer to another agency unit where his skills could be better used, but his supervisors said no.
"It's an experience. Not one that I would want people to go through, but it's an experience," he said.
The Health Department said in a statement, "This employee engaged in several acts of professional misconduct that resulted in his suspension." ||||| He knew just how to press his boss’s buttons!
A city Health Department worker was suspended without pay for answering the agency’s help desk phone using a robot voice, city documents show.
Computer specialist Ronald Dillon, 66, was slapped with a 20-day suspension for speaking in a “deliberately robotic fashion” while fielding calls from co-workers and the general public, according to an Office of Administrative Trials and Hearings decision.
Dillon, who answered calls at the department’s IT help desk, was caught using the weird monotone voice at least five times between February 2012 and April 2013, according to the document.
“You-have-reached-the-Help-Desk. This-is-Mr. Dil-lon. How-may-I help-you?” he would over-enunciate.
His robot impression was so convincing that one annoyed caller demanded to “speak to a human” after she thought an “automated system” had hung up on her.
Other callers filed formal complaints — tipping off his boss, Barry Novack, who told Dillon to stop using the phony voice, according to the decision.
Dillon told a judge at a disciplinary hearing that he was simply reading from a script that his boss had asked him to follow.
His thick Brooklyn accent had caused him trouble in the past, so he was trying to speak slower so that callers could understand him better, he told Administrative Law Judge Kara Miller.
He added that he’s not a “people person,” according to the document.
Miller didn’t buy it, ruling that, “He appears to be [a] disgruntled employee who is acting out.”
Dillon’s position involves handling tech problems on two phone lines — one that deals with his fellow Health Department workers and another for members of the public who are having trouble with the Nycmed.com web site.
To prove he was intentionally alienating callers, the Department of Health recorded five examples to present as evidence at the hearing, documents show.
Channeling his inner Siri will cost him around $5,000 in lost wages for the 20-day suspension.
Dillon has worked for the city since 1976 with no trouble, according to his lawyer William Massey.
“They were exceedingly harsh…We’re very disappointed with the decision,” he told the post.
He also claimed that Dillon’s boss hadn’t treated him fairly.
“Novack was bullying Mr. Dillon, belittling him and refusing to train him…Mr. Dillard went from nothing on his record to a 20 day penalty. That is very severe,” he said.
The Health Department initially gave Dillon a 34-day suspension but the judge lowered it the penalty to 20 days on Feb. 14 .
Dillon appealed that decision with the Civil Services Commission on Oct. 2 and lost. |||||
[A]: It may sound pretty funny to answer customer-service calls at an IT help desk in a robot voice ... but Ronald Dillon didn't stop when his boss told him to, and now he's been suspended from his job at New York City's Health Department, without pay, for 20 days. The computer specialist spoke in a "deliberately robotic fashion" on at least five calls between February and April of last year, according to the disciplinary hearing; the judge heard one recording featuring Dillon talking to a customer in a "slow, monotone, and over-enunciated manner." At least one caller was fooled: The customer called back later, telling a Health Department worker that she "hung up when she heard 'the robot' answer the phone because she needed to speak to a human," says the judge's decision. (She also claimed the "automated system" hung up on her, according to the New York Post.) Dillon, who has worked at the department since 1976 but was transferred to the help desk three years ago despite his objections, insisted he was simply reading from a script his boss gave him, and "articulat[ing] each word because he speaks fast and has a Brooklyn accent, which is sometimes difficult to understand," according to the decision. "They objected to the tone of my voice, so I made it atonal," he tells DNAinfo. But the judge found that he was a "disgruntled" employee and didn't like his new boss or new job, noting, "There is a difference between speaking slowly and distinctly and speaking so robotically that callers did not believe that they were speaking to a person." (This guy says Comcast got him fired from his job.)
[Q]: MINNEAPOLIS (WCCO) — More Minnesotans are in custody Saturday for providing support to terrorist groups.
A source tells WCCO that FBI agents arrested at least six people in Minneapolis and San Diego for suspicion of trying to join ISIS. Authorities say efforts of a joint terrorism task force led to the arrests.
The six people taken into custody may be connected to the on-going investigation of Twin Cities residents traveling to Syria to support ISIS over the last year, according to the source.
U.S. Attorney’s Office in Minnesota and the FBI are holding a news conference Monday morning to provide more details.
Authorities say there is no threat to the public safety.
Last fall, 19-year-old Hamza Ahmed was stopped at a New York airport as he tried to travel to Syria. He pleaded not guilty to charges of supporting the terrorism group and lying to the FBI. He’s scheduled to be in court in May. ||||| Updated: 04/21/2015 7:07 AM
Created: 04/20/2015 9:51 AM KSTP.com
Six Minnesota men who appeared in federal court Monday are charged with providing support to a terrorism organization, the U.S. Attorney's Office said.
The men were arrested in Minnesota and California on Sunday.
Nineteen-year-old Zacharia Yusuf Abdurahman, 19-year-old Adnan Farah, 19-year-old Hanad Mustafe Musse and 20-year-old Guled Ali Omar were arrested in Minneapolis. Twenty-one-year-old Adirahman Yasin Daud and 21-year-old Mohamed Abdihamid Farah were arrested in San Diego. All six are Somali-American, the FBI said.
The charges and arrests are tied to the efforts of the six men to try to get to Syria to join the terrorist group Islamic State of Iraq and the Levant (ISIL).
"We have a terror recruiting problem in Minnesota," U.S. Attorney for Minnesota Andrew Luger said at a news conference Monday. "This case demonstrates how difficult it is to put an end to recruiting here."
No evidence exists that men had any plans to conduct any terrorism acts in Minnesota, Luger said.
Luger said the six men charged Monday were making great efforts to try to get to Syria to join the Islamic State, a group that has been terrorizing and gaining ground in Syria, Iraq and other countries.
At least nine Minnesotans have been charged now as part of a conspiracy to provide support to ISIL.
The six men charged Monday are accused of working with Abdi Nur, who is in Syria, the complaint said.
Read the criminal complaint here
"They never stopped plotting another way to get to Syria. These men are focused men who were intent on joining a terrorist organization by any means possible," Luger said.
Luger said the six men are not part of an organized recruiting group but were friends who worked to recruit and help each other get to Syria to support ISIL.
There is not one master ISIL recruiting organizer in Minnesota, Luger said; that makes it more difficult to stop.
"I will work and help anyone who in good faith wants to break the cycle of terror recruiting in Minnesota," Luger said. "The problem will not go away unless we address it head on. It is not a Somali problem. It is not an immigrant problem."
Meanwhile, pilot programs that aim to prevent terrorist threats before they start are being implemented in Minneapolis, Boston, and Los Angeles. Leaders from Minneapolis' Somali community were in Washington, D.C. in February to announce their plan.
"Our pilot program hopefully will be successful because what it's addressing is the underlying cause of some of the problems, which is jobs, lack of opportunities, unemployment, social exclusion," Minneapolis City Council Member Abdi Warsame said at the time.
In other words, if young men and women in Minnesota have social programs and opportunities to turn to, the allure of terrorist organizations will fade.
"Throwing kids in prison did not work," community activist Omar Jamal said outside of the Federal Courthouse in St. Paul Monday, echoing that train of thought. "It didn't prevent this from happening again, so I think there has to be some other ways to handle the kids." ||||| (CNN) A group of six young Minnesota men conspired to sneak into Syria and join ISIS "by any means necessary," federal prosecutors said Monday.
The group of friends, ages 19 to 21, were arrested Sunday.
"What's remarkable about this case was that nothing stopped these defendants from plotting their goal," said U.S. Attorney Andrew M. Luger of the District of Minnesota.
"They were not confused young men. They were not easily influenced. These are focused men who are (were?) intent on joining a terrorist organization by any means possible."
Recruiting for the ISIS terrorist network is a particular problem in Minnesota's community of Somali immigrants.
Read More |||||
[A]: A yearlong FBI investigation into alleged ISIS-related activities resulted in the arrest of six Somali-American men in Minnesota and California yesterday, officials tell CNN. KSTP reports all hail from Minnesota, and it has the US attorney for the state, Andrew Luger, putting the situation in blunt terms: "We have a terror recruiting problem in Minnesota." But there isn't one master recruiter who can be brought down, he says; "this case demonstrates how difficult it is to put an end to recruiting here." KTSP reports three other Minnesotans have previously been charged with trying to aid ISIS, and the AP notes that since 2007, at least 22 men have headed to Somalia from the state to team up with al-Shabab. Authorities say there was no threat to public safety, WCCO reports, with Luger confirming there was no indication that the men wanted to inflict any harm on their home state. He says they were hoping to get to Syria (three were actually stopped at an NYC airport in November, and allegedly lied when they were stopped, the AP notes) and have been charged with conspiracy to provide support to a terror organization. The AP lists the accused as Mohamed Abdihamid Farah, 21; Adnan Abdihamid Farah, 19; Abdurahman Yasin Daud, 21; Zacharia Yusuf Abdurahman, 19; Hanad Mustafe Musse, 19; and Guled Ali Omar, 20. (Meanwhile, American women may be being targeted as ISIS recruits.)
[Q]: PROVIDENCE, R.I. (WPRI) - Rhode Island's House Minority Leader, who's charged with drug possession and driving under the influence , said Monday that he will fight the charges against him.
House Minority Leader Robert Watson (R-East Greenwich-West Greenwich) was arrested Friday after being stopped at a sobriety checkpoint in East Haven, Conn. According to a police report released Monday, Watson was charged with operation while under the influence; possession of drug paraphernalia, a wood marijuana pipe; and possession of a controlled substance, less than four ounces of marijuana.
The arresting officer said he smelled alcohol and marijuana emanating from Watson's body and that Watson was slurring his words. The officer also said Watson failed field sobriety tests.
After placing Watson under arrest, the officer said he found a small plastic bag containing a "green leafy plant-like substance" and a small wooden marijuana smoking pipe.
In a statement, Watson said he travelled to Connecticut to help a friend move. He said he was on his way back home to Rhode Island when he came upon the sobriety checkpoint.
"While confident I was a safe and sober operator, trace evidence of marijuana was discovered and I was charged with operating under the influence, a charge I vehemently deny," Watson said in his statement.
The minority leader said a breathalyzer test showed his blood alcohol level was below the legal limit of .08.
According to the police report, Watson's BAC was .05. Police said he also submitted to a urine test, and the results are still pending. Police said Watson admitted to having one or two drinks, but told officers he was ok to drive.
The representative was at the center of two different controversies so far in 2011. Providence's Guatemalan community became angry in February over comments he made in a speech to the Greater Providence Chamber of Commerce -- questioning the priorities of the General Assembly.
Then, earlier this month, he got into a shouting match on the House floor with House Speaker Gordon Fox over the debate of same sex marriage vs. civil unions. ||||| Drug-bashing RI Republican charged with drug use
By Kase Wickman
Monday, April 25, 2011 11:12 EDT
Robert Watson, a high-ranking Republican state legislator in Rhode Island, is in hot water after being charged with driving under the influence of marijuana and possession of marijuana and drug paraphernalia. Drug charges alone would be bad enough for a public official, but Watson, Rhode Island’s House minority leader, is still remembered for his controversial anti-drug, anti-gay and anti-immigrant remarks.
In February, Watson said the Rhode Island legislature had their priorities right “if you are a Guatemalan gay man who likes to gamble and smokes marijuana.”
Watson was a guest on a radio show soon after the luncheon where he made the misstep, and in response to the outcry over his comments said, “I reject the suggestion that it’s insulting.” He went on to say that lawmakers were “preoccupied with a number of issues, primarily social issues,” when Watson thought they should focus on the economy.
“I apologize when appropriate and/or necessary,” Watson told the Providence Journal in February. “I identify this situation as representing neither circumstance.”
Watson was pulled over at a police checkpoint Friday, East Haven police told the Associated Press. Police noted a “strong odor of marijuana” coming from Watson’s car, and charged him with possession and driving under the influence. The lawmaker was released after he promised to appear in court, and has not yet issued a statement.
Photo courtesy of Rhode Island Legislature. |||||
[A]: | A Rhode Island Republican lawmaker who made a splash with some anti-drug comments a few months ago has been arrested for allegedly driving under the influence of marijuana and possessing drug paraphernalia, Raw Story reports. House Minority Leader Robert Watson drew gasps in February for saying that the state legislature had its priorities straight “if you are a Guatemalan gay man who likes to gamble and smoke marijuana.” Watson’s office confirmed yesterday that he’d been arrested in Connecticut (where he was reportedly helping a friend move), but today he issued a statement saying he intends to fight the charges. “While confident I was a safe and sober operator, trace evidence of marijuana was discovered, and I was charged with operating under the influence, a charge I vehemently deny,” Watson said, according to WPRI. The arresting officer reported smelling alcohol and marijuana when he pulled Watson over, but Watson’s breathalyzer test had him below the legal alcohol limit.
| 5 | NIv2 | task1291_multi_news_summarization | fs_opt | [
0.05200623720884323,
0.5213894844055176,
0.083828866481781,
-0.44360530376434326,
0.6818821430206299,
-0.168133944272995,
0.5284203290939331,
0.6047353744506836,
-0.0242866612970829,
-0.10657024383544922,
0.10236037522554398,
0.47903329133987427,
-0.6378276348114014,
0.1914844661951065,
... |
In this task you are given a tweet. You must judge whether the author of the tweet is angry or not. Label the instances as "Angry" or "Not angry" based on your judgment.
Oh daaaaaaamn @SophiaBush! #getitgirl #girlpower #rawr #ChicagoPD
Not angry
@PuddlesPityP Witnessed #totality, overcome with insane awe & emotion. Son equated to witnessing your talent. #hegetsit #cry #eclipse2017
Not angry
... flat party and I instantly get bollocked about it.
| Angry
| 0 | NIv2 | task397_semeval_2018_task1_tweet_anger_detection | fs_opt | [
-0.7749687433242798,
0.48217475414276123,
0.29225996136665344,
-0.5426124334335327,
-0.35187333822250366,
-1.0446240901947021,
0.2855291962623596,
0.5972971320152283,
0.8430020809173584,
0.3375398516654968,
0.4123227298259735,
-0.42727863788604736,
-0.29951271414756775,
-0.5897229909896851... |
You will be given a definition of a task first, then some input of the task.
You will be given a context, a subject and a relation. Your task is to generate a question based on the subject and relation. The generated question should include the given subject. Try to use a minimum number of words that are not present in either context, subject or relation while generating question.
Context : British Airways Flight 2276 was a scheduled passenger flight which caught fire during take-off from Las Vegas-McCarran International Airport on 8 September 2015, prompting an aborted take-off and the evacuation of all passengers and crew.
Subject : British Airways Flight 2276
Relation : point in time
Output: | What date is associated with British Airways Flight 2276? | 1 | NIv2 | task1325_qa_zre_question_generation_on_subject_relation | zs_opt | [
0.1302778124809265,
0.33832621574401855,
-0.0870167464017868,
-0.28488144278526306,
0.2758706212043762,
0.7002382874488831,
0.10168145596981049,
0.23035714030265808,
-0.2044053077697754,
0.3073735237121582,
-0.39737021923065186,
-0.255577027797699,
-0.32036077976226807,
0.21797488629817963... |
Detailed Instructions: In this task you are given a list of integers and you need to find the absolute value of the difference between each two consecutive values. The output should be a list of the absolute value of the differences of each two consecutive values.
Q: [-11, 30, -6, -79, -5, 23, -89, -100, -71, 63, -68, 83, 25]
A: | [41, 36, 73, 74, 28, 112, 11, 29, 134, 131, 151, 58] | 9 | NIv2 | task125_conala_pair_differences | zs_opt | [
-0.4371206760406494,
-0.14296932518482208,
-0.07824122160673141,
-0.7937623262405396,
-0.35869288444519043,
-0.6028959155082703,
1.4551023244857788,
0.036805517971515656,
0.23418250679969788,
-0.3611815571784973,
0.05421604961156845,
-0.2782343626022339,
-0.19482439756393433,
-0.0675725936... |
Detailed Instructions: In this task, you are given an input list A. You need to find all the elements of the list that are numbers in the same order as they appear in the list A.
Problem:['983', '587', '7005', '9417', '2879', 'o', 'v', 'U', 'z', '9479', '7401', 'h', 'V', 'T', '1127', 'T', '1929', 'N']
Solution: | 983, 587, 7005, 9417, 2879, 9479, 7401, 1127, 1929 | 8 | NIv2 | task497_extract_all_numbers_from_list_in_order | zs_opt | [
-0.5264331102371216,
0.7234179377555847,
-1.1024649143218994,
-0.13481594622135162,
-0.3059965968132019,
-0.20426654815673828,
0.5403052568435669,
-0.42742711305618286,
-0.2362731695175171,
0.09692347049713135,
-0.6967769265174866,
-0.6561899781227112,
0.07802025973796844,
0.20815165340900... |
In this task, you are given a short story consisting of exactly 5 sentences where the second sentence is missing. You are given a candidate for the second sentence and you need to identify if the given sentence connects the first sentence with the rest of the story. Indicate your answer by "Yes" if it connects, otherwise "No". Do not generate anything else apart from "Yes" or "No". The given sentence is incorrect if it changes the subsequent storyline, so that at least one of the three subsequent sentences form a consistent story.
Q: Sentence 1: Alice loved apple pie.
Sentence 3: Alice decided she would look for a recipe to bake a pie
Sentence 4: Alice found her grandmother's recipe
Sentence 5: Alice decided she would use her grandmother's recipe to bake a pie
Given Sentence 2: However, she did not know how to bake a pie.
A: Yes
****
Q: Sentence 1: It was Jon's last game of the season.
Sentence 3: It was Jon's turn up to bat
Sentence 4: Jon saw the pitch and swung
Sentence 5: The ball sailed over the fence and Jon won the game!
Given Sentence 2: His team was losing by one point.
A: Yes
****
Q: Sentence 1: Tanner was an actor in a play.
Sentence 3: Unfortunately, his sister was sick and he was held up
Sentence 4: Finally she was able to get to sleep and he left for the show
Sentence 5: When he got there, he was too late and they had sent in an understudy
Given Sentence 2: He decided to take a pottery class.
A: | No
****
| 4 | NIv2 | task066_timetravel_binary_consistency_classification | fs_opt | [
-0.13369925320148468,
0.39384785294532776,
-0.42938411235809326,
0.6041544675827026,
-0.04825875163078308,
-0.9682131409645081,
0.49401533603668213,
1.1704225540161133,
-0.24745731055736542,
-0.05288364738225937,
-0.3341078758239746,
-0.03595732897520065,
-0.3079124689102173,
0.02526102215... |
In this task, you are given one english sentence. The major part of the sentences talk about health-related topics, but some of the them describe an organization and its activities. Your job is to translate the given sentences into German.
One example is below.
Q: Flu in birds is quite common.
A: Grippe bei Vögeln kommt relativ häufig vor.
Rationale: The translation is correct.
Q: The meetings are usually organized in plenary sessions and workshops, facilitating the discussion and the consensus regarding outcomes.
A: | Die Tagungen sind in der Regel in Form von Plenarsitzungen und Workshops organisiert, was die Diskussion und die Konsensbildung über die Ergebnisse erleichtert. | 9 | NIv2 | task1396_europa_ecdc_tm_en_de_translation | fs_opt | [
-0.3504016697406769,
0.6764262914657593,
-0.30827105045318604,
-0.529733419418335,
-0.28750866651535034,
-0.580280601978302,
0.6246616840362549,
0.2589150071144104,
-0.08941969275474548,
-0.14638270437717438,
-0.10078826546669006,
0.5934036374092102,
-0.6629014015197754,
0.6964151263237,
... |
Q: In this task, you are given the abstract of a research paper. Your task is to generate a summary of this abstract. Your summary should not be very short, but it's better if it's not more than 30 words.
Markov Logic Networks (MLNs), which elegantly combine logic rules and probabilistic graphical models, can be used to address many knowledge graph problems. However, inference in MLN is computationally intensive, making the industrial-scale application of MLN very difficult. In recent years, graph neural networks (GNNs) have emerged as efficient and effective tools for large-scale graph problems. Nevertheless, GNNs do not explicitly incorporate prior logic rules into the models, and may require many labeled examples for a target task. In this paper, we explore the combination of MLNs and GNNs, and use graph neural networks for variational inference in MLN. We propose a GNN variant, named ExpressGNN, which strikes a nice balance between the representation power and the simplicity of the model. Our extensive experiments on several benchmark datasets demonstrate that ExpressGNN leads to effective and efficient probabilistic logic reasoning.
A: | We employ graph neural networks in the variational EM framework for efficient inference and learning of Markov Logic Networks. | 7 | NIv2 | task668_extreme_abstract_summarization | zs_opt | [
-1.363739252090454,
-0.2864135503768921,
0.06816482543945312,
-0.7512730956077576,
0.019055336713790894,
-0.6636676788330078,
0.09302730858325958,
0.40466779470443726,
0.6851152777671814,
-0.09083699434995651,
-1.0882585048675537,
0.32122913002967834,
-0.73105788230896,
0.19267623126506805... |
Teacher:In this task, you're given a short story of five sentences written in natural language. However, the order of the given story is not correct. Your job is to return the correct order for the given five sentences to create a coherent short story with the new order that has the correct flow. Generate your answer using the number of sentences in the correct order, such as '23415'.
Teacher: Now, understand the problem? Solve this instance: Sentence1: It was in a department store. Sentence2: Eventually the horned toad died from starvation. Sentence3: The toad looked very fierce. Sentence4: We tried to feed it but it would not eat. Sentence5: I bought a horned toad when I was ten.
Student: | 25341 | 6 | NIv2 | task300_storycloze_order_generation | zs_opt | [
-0.08897188305854797,
0.2093086540699005,
-0.24051479995250702,
-1.0530532598495483,
-0.25744813680648804,
-0.46158838272094727,
-0.03681137040257454,
0.6124292016029358,
-0.45719748735427856,
0.17475038766860962,
-0.5850565433502197,
-1.067387342453003,
-1.2536715269088745,
0.121582463383... |
You will be given a definition of a task first, then an example. Follow the example to solve a new instance of the task.
In this task, you will be presented with a question about part-of-speech tag of a word in the question. You should write the required POS tag answering the question. Here is the Alphabetical list of part-of-speech tags used in this task: CC: Coordinating conjunction, CD: Cardinal number, DT: Determiner, EX: Existential there, FW: Foreign word, IN: Preposition or subordinating conjunction, JJ: Adjective, JJR: Adjective, comparative, JJS: Adjective, superlative, LS: List item marker, MD: Modal, NN: Noun, singular or mass, NNS: Noun, plural, NNP: Proper noun, singular, NNPS: Proper noun, plural, PDT: Predeterminer, POS: Possessive ending, PRP: Personal pronoun, PRP$: Possessive pronoun, RB: Adverb, RBR: Adverb, comparative, RBS: Adverb, superlative, RP: Particle, SYM: Symbol, TO: to, UH: Interjection, VB: Verb, base form, VBD: Verb, past tense, VBG: Verb, gerund or present participle, VBN: Verb, past participle, VBP: Verb, non-3rd person singular present, VBZ: Verb, 3rd person singular present, WDT: Wh-determiner, WP: Wh-pronoun, WP$: Possessive wh-pronoun, WRB: Wh-adverb
What is the part-of-speech tag of the word "the" in the following question: Who were the builders of the mosque in Herat with fire temples ?
Solution: DT
Why? This is a good example. POS tag of "the" is DT.
New input: What is the part-of-speech tag of the word "What" in the following question: What year was the building built with date listed # 94000865 ?
Solution: | WP | 0 | NIv2 | task382_hybridqa_answer_generation | fs_opt | [
-0.03629710152745247,
0.7343032360076904,
-0.3165163993835449,
0.6728142499923706,
-0.06801126897335052,
-0.44123926758766174,
0.442766398191452,
0.5816465020179749,
-0.5528016686439514,
0.23734696209430695,
-0.2640315592288971,
0.13076993823051453,
-0.14824560284614563,
0.0197997428476810... |
Detailed Instructions: In this task, you're given a pair of sentences, sentence 1 and sentence 2, that agree with each other. Your job is to alter sentence 2 so that the pair contradict each other. Generated sentences must be short, with less than 15 words. New information can be introduced. Avoid using pronouns to confuse the subject of the sentence.
Problem:Sentence 1: A man and a woman pose for the camera with pursed lips. Sentence 2: A man and a woman have their photo taken.
Solution: | A man and a woman run away from the camera. | 8 | NIv2 | task187_snli_entailment_to_contradiction_text_modification | zs_opt | [
-0.6022966504096985,
0.8419809937477112,
-0.18249323964118958,
-0.8478183746337891,
0.05986028164625168,
0.30998849868774414,
0.19049806892871857,
0.40840622782707214,
0.19428572058677673,
-0.745489776134491,
-0.9489842653274536,
-0.7047784328460693,
-0.17010091245174408,
0.238470017910003... |
Q: You are given a sentence in English. Your job is to translate the English sentence into Hebrew.
It would be a personalized chip just for you.
A: | זה היה שבב שעשוי לפי מידותייך בדיוק. | 7 | NIv2 | task1221_ted_translation_en_he | zs_opt | [
-0.6747696399688721,
0.4237038791179657,
-0.2742897868156433,
0.09164804965257645,
-0.5698931813240051,
-1.1232807636260986,
0.27770864963531494,
-0.16492514312267303,
1.1380386352539062,
0.0025443716440349817,
0.37229421734809875,
0.7410572171211243,
-0.6549854874610901,
0.238222002983093... |
In this task, you need to provide the parts-of-speech tag of a word present in a sentence specified within curly braces ( '{{ ... }}' ). The parts-of-speech tags are fine labels that represent a category of words with similar grammatical properties. The list of part-of-speech tags i.e tagset of this corpus is : '$': Dollar Sign, "''": Single Quotes, ',': Comma Symbol, '-LRB-': Left Parantheses, '-RRB-': Right Parantheses, '.': Period, ':': Colon, 'ADD': Email Address, 'AFX': Affix, 'CC': Coordinating conjunction, 'CD': Cardinal Number, 'DT': Determiner, 'EX': Existential there, 'FW': Foreign Word, 'GW': Go with, 'HYPH': Hyphen symbol, 'IN': Preposition or a subordinating conjunction, 'JJ': Adjective, 'JJR': A comparative Adjective, 'JJS': A Superlative Adjective, 'LS': List item Marker, 'MD': Modal, 'NFP': Superfluous punctuation, 'NN': Singular Noun, 'NNP': Singular Proper Noun, 'NNPS': Prural Proper Noun, 'NNS': Prural Noun, 'PDT': Pre-determiner, 'POS': Possessive Ending, 'PRP': Personal pronoun, 'PRP$': Possessive Pronoun, 'RB': Adverb, 'RBR': Comparative Adverb, 'RBS': Superlative Adverb, 'RP': Particle, 'SYM': Symbol, 'TO': To , 'UH': Interjection, 'VB': Base form Verb, 'VBD': Verb in Past tense, 'VBG': Verb in present participle, 'VBN': Verb in past participle, 'VBP': Verb in non-3rd person singular present, 'VBZ': Verb in 3rd person singular present, 'WDT': Wh-determiner, 'WP': Wh-pronoun, 'WP$' Possessive Wh-pronoun, 'WRB': Wh-adverb, 'XX': Unknown, '``': Double backticks.
Input: Consider Input: Sentence: You ca n't go {{ wrong }} with Tuesday prices , even if you get quite the mixed bag of comedians !
Word: wrong
Output: RB
Input: Consider Input: Sentence: Never before has {{ such }} a convenient method been available for consuming nutritional products .
Word: such
Output: PDT
Input: Consider Input: Sentence: Otherwise , we can see who we can scare up at the office {{ , }} or you can have Laura sign and I 'll initial later .
Word: ,
| Output: ,
| 2 | NIv2 | task584_udeps_eng_fine_pos_tagging | fs_opt | [
0.06708037853240967,
0.31067514419555664,
0.009210558608174324,
0.18226802349090576,
0.0017046872526407242,
-0.5464112162590027,
0.6485403180122375,
0.9836719632148743,
-0.5103045701980591,
-0.3571033775806427,
-0.1768428385257721,
0.29143810272216797,
-0.30923745036125183,
-0.037673160433... |
TASK DEFINITION: In this task, you're given the title of a story consisting of five sentences, numbered 1 through 5. Your job is to determine which two sentences need to be swapped sentences in order to make a story that makes complete sense and is befittingly titled. Indicate your answer using the numbers of the two sentences in order, such as '34' or '25'. The first digit refers to the sentence which should come first in the story.
PROBLEM: Title: Missed the Bus. Sentence 1: Sara was going to the grocery store on the bus. Sentence 2: She was running late and hurrying towards the bus stop. Sentence 3: Luckily she caught up with another later bus stop. Sentence 4: At first she wanted to give up but she decided to walk. Sentence 5: As she approached she saw the bus arrive, pass her and keep driving.
SOLUTION: 35
PROBLEM: Title: Travis lost his friend. Sentence 1: Travis had a dog. Sentence 2: Travis's dog got sick. Sentence 3: Travis decided it was time to put him down. Sentence 4: Travis lost one of his best friends that day. Sentence 5: So he went to put him down at the vet.
SOLUTION: 54
PROBLEM: Title: Cake. Sentence 1: Trudy wanted some cake. Sentence 2: She bought mix at the store. Sentence 3: She mixed the ingredients. Sentence 4: It came out just perfect! Sentence 5: She put it in the oven.
SOLUTION: | 45
| 8 | NIv2 | task218_rocstories_swap_order_answer_generation | fs_opt | [
0.12328577786684036,
0.26730674505233765,
-0.35255366563796997,
0.13637253642082214,
0.19772914052009583,
-0.0967395007610321,
0.4485488831996918,
0.7153570652008057,
-0.10787779092788696,
0.270879864692688,
-0.7229945659637451,
-0.12792333960533142,
-0.20456302165985107,
0.101181454956531... |
This task is about translating a given English language sentence to Yoruba language.
Example input: While you should not share legal advice to others based on course content, you will develop a high level of expertise upon completion of this course.
Example output: Bí ó ti jẹ́ wí pé o kò leè fún ẹnikẹ́ni nímọ̀ràn ajẹmófin lẹ́yìn tí o bùṣe nínú ẹ̀kọ́ inú ìdá yìí, wà á ní ìmọ̀ kíkún tí kò ní ẹlẹ́gbẹ́ bí o bá parí abala-ẹ̀kọ́ yìí.
Example explanation: This is a good example because the English sentence has been translated correctly to Yoruba.
Q: One does not tie a goat with another goat and keep one from butting the other to death.
A: | A kì í so ẹran mẹ́ran kó kàn án pa. | 3 | NIv2 | task1685_menyo20k_translation | fs_opt | [
-0.39946842193603516,
0.28708475828170776,
0.3659328818321228,
-0.17342399060726166,
-0.44511836767196655,
-0.2848443388938904,
0.11894543468952179,
0.5990837216377258,
-0.34065401554107666,
-1.1137906312942505,
0.15879422426223755,
0.40971189737319946,
-1.3335554599761963,
0.7642059326171... |
Q: You are given a statement written in Gujarati. Choose the most logical word from the given 4 options which can be used to replace the <MASK> token in the statement. Output the word from the correct option .
Statement: ચિરોડા (રાજપરા) <MASK> દેશના પશ્ચિમ ભાગમાં આવેલા ગુજરાત રાજ્યના સૌરાષ્ટ્ર વિસ્તારમાં આવેલા સુરેન્દ્રનગર જિલ્લામાં આવેલા કુલ ૧૦ (દસ) તાલુકાઓ પૈકીના એક એવા ચોટીલા તાલુકામાં આવેલું એક ગામ છે. ચિરોડા (રાજપરા) ગામના લોકોનો મુખ્ય વ્યવસાય ખેતી, ખેતમજૂરી તેમ જ પશુપાલન છે. આ ગામમાં મુખ્યત્વે ઘઉં, જીરુ, વરિયાળી, બાજરી, કપાસ, દિવેલા, રજકો તેમ જ અન્ય શાકભાજીના પાકની ખેતી કરવામાં આવે છે. આ ગામમાં પ્રાથમિક શાળા, પંચાયતઘર, આંગણવાડી તેમ જ દૂધની ડેરી જેવી સવલતો પ્રાપ્ય થયેલી છે.
Option A: આંગણવાડી
Option B: ભારત
Option C: સૌરાષ્ટ્ર
Option D: પંચાયતઘર
A: | ભારત | 7 | NIv2 | task946_wiki_cloze_gu_multiple_choice_question_answering | zs_opt | [
0.21160626411437988,
0.882124662399292,
0.22906364500522614,
0.006337238010019064,
-0.3104979991912842,
-0.0808468908071518,
-0.021667033433914185,
1.1430028676986694,
-0.06710522621870041,
-0.10679936408996582,
-0.1448689103126526,
0.11372475326061249,
-0.464020311832428,
-0.3494474887847... |
This task involves creating questions from a given passage that involve some kind of complex reasoning (including numerical reasoning).
The generated questions must require looking at more than one part of the passage to answer. Try to use a variety of reasoning types in your questions (some of the sample reasoning types are illustrated in the 'positive examples' field) and also have a variety of answer types (spans, numbers, dates). A span is a continuous phrase taken directly from the passage or question. In such questions, the answer spans are recommended to be five words. Questions with answer type "number", are expected to start with "How many". For instance, "How long did an event last?" can be phrased as "How many years did an event last?". To increase diversity, try to create more questions for which the answer will be multiple spans. Please create questions that require AT LEAST 2 arithmetic operations as illustrated in the 'positive examples' field. You are encouraged to try DIFFERENT COMBINATIONS of operations and not just limit to the combinations shown in those examples. The more challenging questions you create, the better it is.
Ex Input:
Passage: Following the high-scoring loss to the Packers, the Vikings, continuing with Christian Ponder as their quarterback, travel to Arlington, Texas to face the Dallas Cowboys. The first quarter was all field goals, as Dallas scored first with a 41-yarder by Dan Bailey, followed by a 23-yarder for the Vikings by Blair Walsh. In the second quarter, Bailey made another field goal from 44 yards to put the Cowboys up by 3, but the Vikings took a 10-6 lead into halftime, capping the ensuing 79-yard drive with a 6-yard run by Ponder. After the half, Cowboy quarterback Tony Romo completed two consecutive 26-yard passes to tight end Jason Witten to give them an early touchdown and restore their three-point lead. On the very next play from scrimmage, Ponder fumbled the ball as he was sacked in the end zone, and the Cowboys' Nick Hayden recovered it for a touchdown, meaning the Vikings went from 10-6 up to 20-10 down in the space of two scrimmage plays. The Vikings responded immediately with a quick, six-play drive, culminating with a 31-yard pass from Ponder to Kyle Rudolph to cut the Cowboys' lead back to three points. After a series of punts going into the fourth quarter, Adrian Peterson scored for the Vikings with an 11-yard run with almost six minutes to go, but Walsh pushed the extra point kick wide right, giving the Vikings a three-point lead over the Cowboys. Vikings cornerback A. J. Jefferson intercepted Romo on the Cowboys' ensuing drive, putting the Vikings in a position to potentially take over the game, but the offense went three-and-out and was forced to punt. The Cowboys progressed downfield quickly, never faced with a third down, and scored with a 7-yard touchdown pass from Romo to Dwayne Harris, the third time the Vikings had given up a game-winning score in 2013. With less than 30 seconds to play, but they were unable to make any significant territorial gains and Ponder's last-second hail mary fell short, giving the Cowboys a 27-23 win.
Ex Output:
How many field goals did Dan Bailey kick?
Ex Input:
Passage: Following World War II, Britain retained control of both British Somaliland and Italian Somaliland as protectorates. In 1950, as a result of the Paris Peace Treaties, the United Nations granted Italy trusteeship of Italian Somaliland, but only under close supervision and on the condition—first proposed by the Somali Youth League and other nascent Somali political organizations, such as Hizbia Digil Mirifle Somali and the Somali National League —that Somalia achieve independence within ten years. British Somaliland remained a protectorate of Britain until 1960. In 1948, under pressure from their World War II allies and to the dismay of the Somalis, the British returned the Haud and the Ogaden to Ethiopia, based on a treaty they signed in 1897 in which the British, French and Italians agreed upon the territorial boundaries of Ethiopia with the Ethiopian Emperor Menelik in exchange for his help against raids by hostile clans. Britain included the provision that the Somali residents would retain their autonomy, but Ethiopia immediately claimed sovereignty over the area. This prompted an unsuccessful bid by Britain in 1956 to buy back the Somali lands it had turned over. Britain also granted administration of the almost exclusively Somali-inhabited Northern Frontier District to Kenyan nationalists despite an informal plebiscite demonstrating the overwhelming desire of the region's population to join the newly formed Somali Republic.
Ex Output:
British Somaliland remained a protectorate of Britain for how many years after the Paris Peace Treaties?
Ex Input:
Passage: The Rangers started off the 2015–16 season well. After 18 games, they had a 14–2–2 record and a nine-game winning streak, which was eventually halted by Tampa Bay. However, the organization soon went downhill in the winter, losing three-straight games against the Montreal Canadiens, Boston Bruins and Philadelphia Flyers. They eventually posted a 4–7–2 record in December for only ten points. In January the Rangers started playing more efficiently, posting a fair 6–4–1 record, and improved in February by going on a 10–3–1 run without any back-to-back losses. On January 8, 2016 the Rangers traded Emerson Etem to the Vancouver Canucks for Nicklas Jensen and a sixth-round pick in the 2017 NHL draft. On February 28, the Rangers traded prospect Aleksi Saarela, a 2016 second-round draft pick, and a 2017 second-round draft pick for Carolina Hurricanes captain Eric Staal, who later left the team after the season. The Rangers finished the season with 101 points for back-to-back 100+ point seasons. Despite high hopes, the Rangers were eliminated by the Pittsburgh Penguins in the first round of the 2016 Stanley Cup playoffs in five games. On May 2, the Rangers agreed to terms with Antti Raanta on a contract extension, and on May 13, signed Pavel Buchnevich to an entry-level contract.
Ex Output:
| How many days after the Rangers agreed to terms with Antti Raanta on a contract extension did they sign Pavel Buchnevich to an entry-level contract?
| 1 | NIv2 | task026_drop_question_generation | fs_opt | [
0.8716995120048523,
-0.041435353457927704,
-0.5637562274932861,
-0.13320830464363098,
0.17503201961517334,
-0.12829536199569702,
1.0106265544891357,
0.6247004270553589,
0.08708572387695312,
0.1399296522140503,
-0.35816824436187744,
1.0244581699371338,
-0.8692688941955566,
0.575137376785278... |
Detailed Instructions: In this task, you're given a context, a question, three options, and an answer. Your task is to classify whether the given answer is correct or not by providing 'Yes' or 'No', based on the context with commonsense reasoning about social situations.
Q: Context: Jordan, who eats clean, lives like a minimalist and conserves paper to save the world is also a political activist.
Question: How would you describe Jordan?
Options: (A) proud (B) known for his knowledge on how to eat clean (C) not the kind of person to be a vegetarian
Answer: B
A: | Yes | 9 | NIv2 | task579_socialiqa_classification | zs_opt | [
-0.47155287861824036,
-0.006881690584123135,
0.5161746740341187,
-0.30159276723861694,
-0.7179470658302307,
-0.9110674858093262,
0.5059389472007751,
0.9362739324569702,
0.3549371361732483,
0.6273446083068848,
0.048964690417051315,
0.10773216187953949,
-0.8337306976318359,
-0.22111944854259... |
Instructions: You will be given a context and a question in Spanish. Your job is to generate answers that are at least THREE words long.
The answers need to be context specific and can not be general knowledge or a random guess.
Input: CONTEXT: Ignacio de Antioquía (en griego: Ἰγνάτιος Ἀντιοχείας) (Siria, Imperio romano, 35 - Roma, entre 108 y 110) es uno de los padres de la Iglesia y, más concretamente, uno de los padres apostólicos por su cercanía cronológica con el tiempo de los apóstoles. Fue el primero en llamar Católica a la Iglesia. Es autor de siete cartas que redactó en el transcurso de unas pocas semanas, mientras era conducido desde Siria a Roma para ser ejecutado o, como él mismo escribió:
QUESTION: ¿Cuál fue el papel de Ignacio de Antioquía en los inicios del catolicismo?
Output: | es uno de los padres de la Iglesia y, más concretamente, uno de los padres apostólicos por su cercanía cronológica con el tiempo de los apóstoles | 3 | NIv2 | task1334_sqac_answer_generation | zs_opt | [
0.8605900406837463,
0.7007927894592285,
-0.23109187185764313,
-0.14750508964061737,
-0.6408364772796631,
-0.4818657338619232,
-0.36869847774505615,
0.755342960357666,
-0.2127421647310257,
-0.029655171558260918,
0.348664790391922,
0.25836342573165894,
-0.8556243181228638,
0.0869443565607070... |
In this task, you will be shown a correct English sentence. You need to generate an incorrect form of the input sentence.
Input: Consider Input: Facts are learned from experiences .
Output: Facts are learned by experinces .
Input: Consider Input: Finally , the third reason is that that birds use a type of internal compass ; birds have crystals of the mineral magnetite embedded in their beaks .
Output: finally the third cause that birds use a type of internal compass , birds have crystals of the mineral magetite embedded in their break .
Input: Consider Input: I the same other people see many tims on TV rich people are intrested in politica , sports new technologies and also in spaise .
| Output: I the same other people see many tims on TV rich people are intrested in politica , sports new technologies and also in spaise .
| 2 | NIv2 | task1558_jfleg_incorrect_answer_generation | fs_opt | [
0.41134005784988403,
0.6517525911331177,
-0.6339882612228394,
-0.4366799592971802,
0.14448867738246918,
-0.6312364339828491,
-0.22042806446552277,
0.4712565541267395,
-0.19580549001693726,
-0.025360584259033203,
-0.524095356464386,
0.3558017611503601,
-0.5861847400665283,
-0.58768153190612... |
In this task, you are given a sentence in the Swedish language and a corresponding English translation of the Swedish sentence. Your task is to generate a label "Yes" if the translation is correct, otherwise generate label "No".
--------
Question: Swedish: Vi hoppas att detta skall inträffa, åtminstone när det gäller finansieringen.
English: Greece still has a large fiscal deficit which it has to finance.
Answer: No
Question: Swedish: Det är medlemsländernas parlament, eller folk i folkomröstningar, som skall styra fördragets utveckling.
English: Let us, however, ask this question: what if we settle these immigrants in Europe?
Answer: No
Question: Swedish: Ett förslag kan läggas, ett förslag bör läggas, på grundval av artikel 280 i fördraget senast vid halvårsskiftet.
English: Elmar Brok referred to very important areas in external relations - the Eastern Partnership, the enlargement process - where we also need to work for credibility on both sides.
Answer: | No
| 7 | NIv2 | task314_europarl_sv-en_classification | fs_opt | [
-0.48223841190338135,
0.664257824420929,
0.6476934552192688,
-0.10157837718725204,
0.14558054506778717,
-0.5675933957099915,
1.3426628112792969,
0.20048600435256958,
0.297907292842865,
-0.16164901852607727,
0.2840261459350586,
1.169007658958435,
-0.8128325939178467,
0.04812588170170784,
... |
Q: Given an trivia question precisely answer the question with a word/phrase/name. External resources such as Wikipedia could be used to obtain the facts.
Which motor manufacturer made the Terios?
A: | daihatsu | 7 | NIv2 | task898_freebase_qa_answer_generation | zs_opt | [
0.6268984079360962,
0.6454205513000488,
0.253337562084198,
0.3327656090259552,
-1.0203443765640259,
-0.5934003591537476,
0.275628924369812,
0.3830118179321289,
0.3233901560306549,
-0.05984177440404892,
-0.3078640401363373,
0.26571550965309143,
-0.10928872227668762,
0.22129669785499573,
-... |
You are given a science question (easy-level) and four answer options (associated with "A", "B", "C", "D"). Your task is to find the correct answer based on scientific facts, knowledge, and reasoning. Do not generate anything else apart from one of the following characters: 'A', 'B, 'C', 'D'. There is only one correct answer for each question.
Which of these is a renewable resource found in Nevada?
(A) zinc metal (B) gravel deposits (C) petroleum products (D) geothermal energy | D | 0 | NIv2 | task228_arc_answer_generation_easy | zs_opt | [
0.4640241265296936,
0.7519564628601074,
-0.3839530348777771,
-0.3494792878627777,
-0.5588764548301697,
-1.0731587409973145,
1.099603295326233,
0.9018118381500244,
0.3547975718975067,
-0.23685282468795776,
-0.38655346632003784,
0.2374783158302307,
-0.097214475274086,
-0.6204222440719604,
... |
Given the task definition and input, reply with output. You are given a passage. Using the information present in the passage, you need to classify it into one of the 10 topics: 0 - 'Society & Culture', 1 - 'Science & Mathematics', 2 - 'Health', 3 - 'Education & Reference', 4 - 'Computers & Internet', 5 - 'Sports', 6 - 'Business & Finance', 7 - 'Entertainment & Music', 8 - 'Family & Relationships', 9 - 'Politics & Government'.
You see, one of the biggest value an MBA will bring is the "networking" opportunity. Hence, a part-time MBA with no classroom-based activities will miss this important piece.\n\nIf you want to go for a fully remote-based degree, the Open university is a good option.\n\nColumbia also has a good option for a certificate: you keep taking several courses and after some time you can apply for a certificate in business excellence.\n\nI think both the part-time mba or the certificate are good options if you pick a good school.\n\nIf you can afford going to the campus about twice a month, an executive mba is the best option in my opinion. Chech NYU, Kellog or Columbia.
| 3 | 5 | NIv2 | task1593_yahoo_answers_topics_classification | zs_opt | [
0.025134362280368805,
0.3421142101287842,
0.006802796386182308,
-0.35959094762802124,
-0.39986810088157654,
0.36469748616218567,
0.4102281928062439,
0.32061856985092163,
0.7765904664993286,
-0.029896559193730354,
0.15680642426013947,
0.41738370060920715,
-0.3379543423652649,
0.082599148154... |
Teacher:Given a prompt and two completions from a language model, select the completion that is more topical with respect to continuing the prompt. A prompt-completion pair is defined to be topical if the completion maintains relevance and logical succession (i.e. stays on topic) with the prompt. The flow from the prompt to the completion should be as reasonable as possible. Generate either Completion A or Completion B.
Teacher: Now, understand the problem? Solve this instance: Prompt: yea but I can't see 11 months making it, NOW u need
Completion A: to be 100% on your side!""You can wait that long if not for me." —Sasuke Yamamoto to Asada
Completion B: to be willing to sacrifice some stuff. So theres no real incentive for anyone to stick with it for a year but there's still som
Student: | Completion B | 6 | NIv2 | task139_detoxifying-lms_classification_topicality | zs_opt | [
-0.34529638290405273,
0.0034732799977064133,
0.46661263704299927,
-0.5335291624069214,
-0.16042156517505646,
-1.1613399982452393,
1.0956279039382935,
0.26187705993652344,
0.19369164109230042,
-0.37889528274536133,
-0.0785364955663681,
0.20061293244361877,
-0.027022819966077805,
0.482895672... |
Teacher:You are given a background paragraph that describes one or more causal or qualitative relationships such as a relationship in economics or a scientific law and a story that makes use of the concepts or the relationship described in the provided paragraph. You are also given a question about the story that requires an understanding of the relationship described in the background paragraph and the story. You need to come up with an answer to the given question; the answer will be a span from either the question or the story. In order to correctly answer the given question, you need to understand the relationship mentioned in the background paragraph and should be able to use it to understand that in the story. Your answer can not consist of any word that is not mentioned in any of these: the background paragraph, the story, or the question. You can directly copy and paste a span from the story or the question while answering the given question.
Teacher: Now, understand the problem? Solve this instance: Background Paragraph: During May, June, and July, the Northern Hemisphere is exposed to more direct sunlight because the hemisphere faces the Sun. The same is true of the Southern Hemisphere in November, December, and January. It is Earth's axial tilt that causes the Sun to be higher in the sky during the summer months, which increases the solar flux. However, due to seasonal lag, June, July, and August are the warmest months in the Northern Hemisphere while December, January, and February are the warmest months in the Southern Hemisphere.
Story: Two twin brothers live far away from each other. Dan lives in the Norther Hemisphere, in North America, and Bill lives in Tasmania, in the Southern Hemisphere.They often talk about the weather, the seasons, and how it affects them.
Question: Which brother experiences less sunlight in August?
Student: | Bill. | 6 | NIv2 | task061_ropes_answer_generation | zs_opt | [
0.5892195105552673,
-0.3893424868583679,
-0.20516493916511536,
-0.1939803659915924,
-0.25485479831695557,
-1.448498249053955,
-0.2438018023967743,
0.5852382779121399,
-0.18224647641181946,
-0.13090187311172485,
-0.1735643893480301,
-0.2619609832763672,
-0.5032879710197449,
-0.1145187467336... |
In this task, you are given a text of article and corresponding title of an article. Your task is to generate label "yes" if headline (title) is right for article, otherwise generate "no".
Example: Article: australia 's current account deficit shrunk by a record #.## billion dollars -lrb- #.## billion us -rrb- in the june quarter due to soaring commodity prices , figures released monday showed . Title: australian current account deficit narrows sharply
Example solution: yes
Example explanation: This headline is appropriate for the given article text because the focus of this article is australian current accounts. Hence, the label is 'yes'.
Problem: Article: us defense secretary robert gates met us and iraqi commanders here monday , reviewing what he said will be `` a mission transition '' as the us force shrinks and turns over more of the country to iraqis . Title: bulgaria arrests human traffickers iraqis at turkish border
| Solution: no | 5 | NIv2 | task289_gigaword_summarization | fs_opt | [
-0.27434614300727844,
0.2715933918952942,
-0.289985328912735,
0.3697856664657593,
-0.3531930446624756,
-0.33310845494270325,
0.41363662481307983,
0.4328444004058838,
0.011468222364783287,
0.2425360530614853,
-0.20294561982154846,
0.14983347058296204,
-0.5572938919067383,
0.3876367211341858... |
You will be given a definition of a task first, then some input of the task.
In this task, you are given inputs i,j, and A, where i and j are integers and A is a list. You need to return the sum of all the numerical elements in the list A between the positions i and j (including positions i and j). Return 0 if no numerical element is present in the list between the given ranges. i and j will be non-negative, and will always have a value less than the length of A. i will always be less than j.
7, 11, ['9599', '2147', '3991', 'p', 'X', '7413', '2837', 'v', 'h', 'U', '6235', '9879', 'E']
Output: | 9072 | 1 | NIv2 | task606_sum_of_all_numbers_in_list_between_positions_i_and_j | zs_opt | [
-0.3627375364303589,
0.1563674360513687,
-0.47888675332069397,
-0.3661208152770996,
0.11752410233020782,
0.5740421414375305,
0.2823737859725952,
-0.03997373580932617,
-0.15234464406967163,
-0.1801050454378128,
-0.7923027873039246,
0.26299118995666504,
0.5266892313957214,
-0.397550135850906... |
Definition: In this task, you need to count the occurrences of the given word in the given sentence.
Input: Sentence: 'a bald headed man cutting a ribbon in front of people on bicycles'. Count the occurrences of the word 'man' in the given sentence.
Output: | 1 | 2 | NIv2 | task158_count_frequency_of_words | zs_opt | [
0.5282950401306152,
1.096043586730957,
-0.3675311803817749,
-0.844579815864563,
-0.5542528629302979,
-0.12564465403556824,
0.8841664791107178,
0.4548702836036682,
-0.19742855429649353,
0.42801564931869507,
-1.2809193134307861,
-0.6026132106781006,
-0.534925103187561,
1.014851450920105,
0... |
Instructions: In this task, you are given a sentence in the English language from the various articles. Your task is to translate the given English sentence into the Yoruba language. Please bear in mind the following guidelines while doing the translation: 1) Generated output should have natural language and formal form. The output sentence should not be a colloquial form of the input sentence. The generated output should be in natural language which is how you would type your queries in a text-based virtual assistant. 2) The words between quotation marks *SHOULD NOT* be translated. The output should keep those values intact and include the quotation marks around them as well. 3) Numbers and fully capitalized words like SEPTEMBER, or 10 HOURS *SHOULD NOT* be translated. Keep them as they are in the translations. 4) Do not localize measurement units like miles to kilometers during your translation. 5) Note the input is in sentence case except for special placeholders. Do the same in your translations.
Input: Religious organizations in Mongolia are required to renew their registrations annually, and our brothers had successfully done this since first being registered in 1999.
Output: | Ìjọba sọ fáwọn ẹlẹ́sìn pé ọdọọdún ni kí wọ́n máa forúkọ ẹ̀sìn wọn sílẹ̀ ní Mòǹgólíà, àtìgbà táwọn ará wa sì ti kọ́kọ́ forúkọ ẹ̀sìn wọn sílẹ̀ lábẹ́ òfin lọ́dún 1999 ni wọ́n ti ń ṣe bẹ́ẹ̀. | 3 | NIv2 | task1619_menyo20k-mt_en_yo_translation | zs_opt | [
-0.41719359159469604,
0.18817344307899475,
0.011711981147527695,
-0.11979201436042786,
0.02826082333922386,
-0.3504731357097626,
0.6523054838180542,
0.38650116324424744,
0.11375238001346588,
-0.463445246219635,
-0.5487634539604187,
-0.3545228838920593,
-0.7691347002983093,
0.30680608749389... |
Teacher: In this task, you are given a multiple-choice question and you have to pick the correct option. Answer with option indexes (i.e., "A", "B", "C", and "D").
Teacher: Now, understand the problem? If you are still confused, see the following example:
Deserts are in extremely dry environments, so liquids will naturally be (A) rainbow (B) plentiful (C) meager (D) talented
Solution: C
Reason: This is a good example and the output correctly answers the question.
Now, solve this instance: Which would directly follow the appearance of grey clouds? (A) UFOs (B) planes (C) sleet (D) tsunami
Student: | C | 2 | NIv2 | task1286_openbookqa_question_answering | fs_opt | [
-0.40535640716552734,
-0.10985833406448364,
-0.26910483837127686,
-0.8368895053863525,
-0.49615973234176636,
-0.5013267993927002,
0.3929210901260376,
1.0239894390106201,
-0.4229689836502075,
0.3533969223499298,
0.12434064596891403,
-0.22013117372989655,
-0.14119231700897217,
-0.48998561501... |
In this task, you need to answer the given multiple-choice question on the gain. Gain is the value by which to multiply the input. Classify your answers into 'a', 'b', 'c', 'd', and 'e'.
Q: Problem: a shopkeeper sold an book offering a discount of 5 % and earned a profit of 23.5 % . what would have been the percentage of profit earned if no discount was offered ?
Options: a ) 140 , b ) 120 , c ) 130 , d ) 110 , e ) 150
A: | c | 4 | NIv2 | task1419_mathqa_gain | zs_opt | [
-0.0851607695221901,
0.3787133991718292,
-0.1270594298839569,
0.011349792592227459,
-0.46862056851387024,
0.033907581120729446,
0.4631993770599365,
0.9666807055473328,
0.15841513872146606,
0.40067043900489807,
-0.979302167892456,
-0.2545219659805298,
-1.474244475364685,
-0.0694299042224884... |
Part 1. Definition
Write an incorrect answer to the given question based on the associated fact. You are also provided with the correct answer to the given question. Make sure that your incorrect answer is relevant and similar to the associated fact. Also, try to make the incorrect answer similar to the correct answer so that distinguishing the correct answer from the incorrect answer is not very easy. Make sure you don't accidentally provide another correct answer! Also, make sure they sound reasonable (e.g., might be on a school pop quiz). A good incorrect answer can be constructed using words associated with the question, but not the correct answer. For example, for the question "What helps plants survive?", using words like "weeds", "vase", "bee" (associated with "plant"), or "first aid", "parachute", "accident" (associated with "survive") etc. Your incorrect answers make the question hard, so these results in good incorrect answers.
Part 2. Example
Fact: pesticides can harm animals.
Question: What can harm animals?
Correct Answer: pesticides.
Answer: pet shampoo.
Explanation: This is clearly an incorrect answer to the question. And since, it is relevant with the given fact, it is a good incorrect answer. Other good incorrect answers are "fresh water" and "clouds".
Part 3. Exercise
Fact: Hail that falls on land may soak into the ground, becoming groundwater.
Question: What might fall on land may soak into the ground, becoming groundwater?
Correct Answer: Hail.
Answer: | Rocks. | 7 | NIv2 | task042_qasc_incorrect_option_generation | fs_opt | [
-0.04006003215909004,
1.1495563983917236,
-0.3369104564189911,
0.11775591969490051,
-0.12919169664382935,
-1.6742563247680664,
0.45501580834388733,
1.1689989566802979,
-0.5506786108016968,
0.3157917261123657,
-1.115678310394287,
0.45779916644096375,
-0.6801255941390991,
-0.1553163826465606... |
Definition: Given a document, find the main entity about whom the author is writing. Write the full name if mentioned in the text. Note that URLs in the text have been replaced with [Link].
Input: -LRB- CNN -RRB- Former Democratic presidential nominee Hillary Clinton , amid the fallout of the massacre in Las Vegas , turned the political spotlight on a gun bill that 's currently making its way through the House .
Clinton repeatedly brought up the need for tougher gun control laws during the 2016 president campaign .
Output: | Hillary Clinton | 2 | NIv2 | task419_persent_answer_generation | zs_opt | [
-0.9636253714561462,
0.433126837015152,
0.08435836434364319,
0.5321794748306274,
0.3539121747016907,
-0.327751487493515,
0.2907402217388153,
0.3857501447200775,
-0.138057678937912,
0.20322638750076294,
0.1179799735546112,
0.14930251240730286,
-0.5896279811859131,
0.12495262920856476,
-0.... |
In this task, you're given a dialogue between a customer and a flight booking agent with a gap in the conversation. Your job is to find the answer of the previous dialogue. Avoid using irrelevant extra information while creating the answer. The answer should be relevant to the question before the blank. If you fill the blank with a question, it should have an answer from the agent in the given dialogue. Fill the gap marked with underline.
Ex Input:
customer: Hello.
__
customer: I am Janet Robert. I want to cancel my reservation due to heavy rainfall at my source code.
agent: Sure, please wait a moment. There is no reservation on your name.
customer: Thank you for informing me.
agent: Thank you so much.
Ex Output:
agent: Hello. How may I assist you?
Ex Input:
customer: Hi.
agent: Hello. How can I help you today?
customer: I need your assistance in book a ticket from JFK to CLT?
agent: Yes, I can assist you. Could you share me your dates of travelling?
customer: My travel dates are 11/02 and 11/04.
agent: Ok, give me a second.
customer: Ok.
agent: Sorry to say that there are no flight services provided on this route.
customer: No problem.
agent: Can I know your name for documentation purpose?
customer: Myself Nicole Mitchell.
__
customer: Thank you for your help.
agent: Happy to serve you anytime.
Ex Output:
agent: Thank you for reaching us.
Ex Input:
customer: Hello.
agent: Hello. How may I assist you at this moment?
customer: I am Jonathan Mitchell from Denver. I want to go to Houston next month, can you please book a flight ticket for me?
agent: Sure, I am glad to help you. Can you please provide your planned dates of travelling?
__
agent: Ok, may I know your origin and destination airport codes for booking?
customer: I want to fly from DEN and destination airport is IAH.
agent: Please wait a moment to check the flights availability.
customer: Ok.
agent: According to your information, I have found a flight with single stop over in your journey with a minimum fare of 200 is provided by Delta airlines. Shall I proceed with this?
customer: I am ok with this flight, please proceed.
agent: Ok, your ticket has been booked in flight number 1002.
customer: Thank you for your help.
agent: Thank you for approaching us, have a great journey.
Ex Output:
| customer: Thanks, I want to take off on 05/31 and book my return ticket also on 06/02.
| 1 | NIv2 | task574_air_dialogue_sentence_generation | fs_opt | [
-0.21653230488300323,
-0.09946133941411972,
-0.9065729379653931,
-0.4442850351333618,
0.32174110412597656,
-0.12324744462966919,
0.8207098841667175,
0.23445168137550354,
-0.4055429697036743,
0.46518224477767944,
0.13622444868087769,
-0.07005240768194199,
-0.5645730495452881,
0.125591099262... |
Given the task definition and input, reply with output. In this task, you are given a review of movie. Your task is to classify given movie review into two categories: 1) positive, and 2) negative based on its content.
Major Payne was really not very good at all. Despite being funny here and there, the story was ridiculous and the acting was poor. Major Payne's voice and temperament were especially annoying. The idea was ridiculous and the things that the boys had to do in that film were even more ridiculous. I would not recommend this film to anyone.
| negative | 5 | NIv2 | task284_imdb_classification | zs_opt | [
-1.0304183959960938,
0.833961009979248,
0.3188081681728363,
-0.03067004680633545,
0.3063735365867615,
0.34236782789230347,
0.4622451066970825,
0.7013434171676636,
-0.017084438353776932,
-0.06969147175550461,
-0.5901251435279846,
-0.42233073711395264,
-0.537022054195404,
-0.6399411559104919... |
In this task, you are given a sentence in English, and your task is to translate it into Persian.
Let me give you an example: I heard enough of what she said to you last night to understand her unwillingness to be acting with a stranger;
The answer to this example can be: دیشب شنیدم به تو چه میگفت. فهمیدم که دوست ندارد با یک غریبه این نقش را بازی کند.
Here is why: This is a good example. The above sentence is correctly translated from English to Persian.
OK. solve this:
As Mr. Rushworth did _not_ come, the injury was increased, and she had not even the relief of shewing her power over him;
Answer: | ولی آقای راشورث نیامد و ناراحتی دوشیزه برترام بیشتر شد، چون حتی این فرصت هم از دست میرفت که لااقل زورش به آقای راشورث برسد. | 8 | NIv2 | task661_mizan_en_fa_translation | fs_opt | [
0.1582699716091156,
1.4795677661895752,
-0.07518210262060165,
-0.1059170812368393,
-0.3691316843032837,
-0.5574466586112976,
1.4042562246322632,
0.3774239122867584,
0.3118547201156616,
-0.07126548141241074,
-0.21349762380123138,
0.4913486838340759,
-0.6664286851882935,
-0.1886187493801117,... |
Detailed Instructions: You are given a sentence in Polish. Your job is to translate the Polish sentence into Portugese.
See one example below:
Problem: Media łączyły się z nami.
Solution: Os "" media "" interligavam-se connosco.
Explanation: The Polish sentence is correctly translated into Portugese, because the meaning is preserved.
Problem: (Śmiech) OK.
Solution: | (Risos) Ok. | 4 | NIv2 | task1264_ted_translation_pl_pt | fs_opt | [
-0.038076646625995636,
0.8006168603897095,
-0.013524942100048065,
-0.9410830736160278,
-0.16773095726966858,
-0.014873540960252285,
0.654388427734375,
-0.15446171164512634,
0.5910912752151489,
-0.1925860494375229,
-0.6073414087295532,
0.6454706192016602,
-0.675946056842804,
-0.039281353354... |
In this task, you're given a statement, the genre to which that statement belongs, and a label indicating if the statement should be agreed with (entailment), disagreed with (contradiction), or neither (neutral). Your job is to write a sentence that describes the genre that follows the tone with respect to the statement, as indicated by the label. If sentence X agrees with sentence Y, the can be concluded from one another. If sentence X disagrees with sentence Y, they can not be correct at the same time. The sentence must also belong to the genre specified.
Statement: you know the season's just so long
Label: contradiction.
Genre: telephone.
You know the season is short and coming to an end already.
Statement: i don't know actually i think it i think it should be a civic level the city level and a a system level really to find out and to to see what they need and not overinflate it
Label: contradiction.
Genre: telephone.
I think it is up to the individual to find out.
Statement: 6For a detailed discussion of our objectives, scope, and methodology, see appendix II.
Label: contradiction.
Genre: government.
| Appendix II has a detailed discussion of sales.
| 0 | NIv2 | task203_mnli_sentence_generation | fs_opt | [
-0.9520305395126343,
0.5222704410552979,
0.05450592562556267,
-0.11640456318855286,
0.16678568720817566,
-0.694632887840271,
-0.1305636167526245,
0.7378536462783813,
-0.009969580918550491,
-0.15804892778396606,
-0.24284113943576813,
-0.10899292677640915,
-0.7765817642211914,
0.237960189580... |
Q: In this task, you're given a passage, further information available on a particular linked term from the statement, and a question. Your job is to generate the answer to the question by using the information provided. If there is no clear answer obtainable, output 'none'.
Passage: York started his career with Handsworth Royal, Birchfield Rangers and the Royal Air Force, and also guested for Chelsea during World War I. In March 1915 he joined Aston Villa as an amateur, signing professional forms in August 1919. He scored one goal in 17 games in 1919–20, but did not feature in the 1920 FA Cup Final, which ended in a 1–0 victory over Huddersfield Town at Stamford Bridge. He appeared just 11 times in 1920–21, before going on to make 47 appearances in the 1921–22 campaign, as the "Villans" finished fifth in the First Division. He scored nine goals in 37 games in 1922–23 and five goals in 43 games in 1923–24. He also appeared at Wembley in the 1924 FA Cup Final, in a 2–0 defeat to Newcastle United. He scored seven goals in 34 matches in 1924–25, before hitting 20 goals in 44 appearances in 1925–26. He bagged 13 goals in 43 games in 1926–27, before being limited to just four goals in 30 appearances in 1927–28. He rediscovered his scoring form with 18 strikes in 48 matches in 1928–29, before hitting seven goals in 32 games in 1929–30. However he played just four times in the 1930–31 campaign, as Villa finished second in the league with an English record of 128 top-flight league goals scored.
Link Information: none Question: Which team had a better win/loss record the year York joined Aston Villa, Huddersfield Town or Aston Villa?
A: | Answer: none | 7 | NIv2 | task238_iirc_answer_from_passage_answer_generation | zs_opt | [
0.45650213956832886,
0.5861597657203674,
-0.18476739525794983,
0.6157614588737488,
0.8386020660400391,
-0.3642923831939697,
0.8548207879066467,
0.8105137944221497,
0.0832517072558403,
-0.25074779987335205,
0.001221796846948564,
0.8774194717407227,
-0.8693279027938843,
0.20776793360710144,
... |
In this task, you will be presented with a question about part-of-speech tag of a word in the question. You should write an implausible POS tag to the question. Even though there exist multiple wrong answers, we only need a single wrong answer. Here is the Alphabetical list of part-of-speech tags used in this task: CC: Coordinating conjunction, CD: Cardinal number, DT: Determiner, EX: Existential there, FW: Foreign word, IN: Preposition or subordinating conjunction, JJ: Adjective, JJR: Adjective, comparative, JJS: Adjective, superlative, LS: List item marker, MD: Modal, NN: Noun, singular or mass, NNS: Noun, plural, NNP: Proper noun, singular, NNPS: Proper noun, plural, PDT: Predeterminer, POS: Possessive ending, PRP: Personal pronoun, PRP$: Possessive pronoun, RB: Adverb, RBR: Adverb, comparative, RBS: Adverb, superlative, RP: Particle, SYM: Symbol, TO: to, UH: Interjection, VB: Verb, base form, VBD: Verb, past tense, VBG: Verb, gerund or present participle, VBN: Verb, past participle, VBP: Verb, non-3rd person singular present, VBZ: Verb, 3rd person singular present, WDT: Wh-determiner, WP: Wh-pronoun, WP$: Possessive wh-pronoun, WRB: Wh-adverb
Example input: What is the part-of-speech tag of the word "the" in the following question: Who were the builders of the mosque in Herat with fire temples ?
Example output: IN
Example explanation: This is a good example. POS tag of the is DT and IN is incorrect.
Q: What is the part-of-speech tag of the word "won" in the following question: For the wrestler who won in 1956 , what was his style of wrestling ?
A: | DT | 3 | NIv2 | task347_hybridqa_incorrect_answer_generation | fs_opt | [
0.1813790500164032,
1.0105892419815063,
-0.41136205196380615,
0.9107033014297485,
0.04041755199432373,
-0.6295700669288635,
0.44011956453323364,
0.6671898365020752,
-0.10782009363174438,
0.06494533270597458,
-0.6217281818389893,
0.5720720291137695,
-0.373919278383255,
-0.18391384184360504,... |
This task is about translating a given English language sentence to French.
Q: Psst!
A: | Psst! | 4 | NIv2 | task1689_qed_amara_translation | zs_opt | [
0.36741191148757935,
0.8960973024368286,
-0.17891016602516174,
-0.701308012008667,
0.3397473692893982,
-0.15947812795639038,
0.5240501165390015,
-0.8966312408447266,
0.9253343343734741,
0.44498708844184875,
-0.007666339632123709,
0.3206186890602112,
0.08997507393360138,
-0.4203178286552429... |
Instructions: In this task you are given a tweet that contains some form of irony. You must classify the type of irony the tweet has. Label the tweets ("polarity","situational","other") based on the irony they have. Situational irony happens when a situation fails to meet some expectations, Label these instances as "situational". polarity irony happens when irony is achieved by inverting the intended sentence, Label these instances as "polarity". There are other kinds of ironies that are neither polarity nor situational, Label these instances as "other". Note that URLs in the text have been replaced with [Link].
Input: Here's the final episode of #minecraft #hardcore survival... We die while trying to get a sustainable food source
Output: | situational | 3 | NIv2 | task387_semeval_2018_task3_irony_classification | zs_opt | [
-1.2527134418487549,
0.34378117322921753,
0.5445178151130676,
-0.3337596654891968,
-0.48625659942626953,
0.12572428584098816,
0.02111693099141121,
0.1876244843006134,
0.1955907940864563,
-0.16749995946884155,
-0.23482945561408997,
-0.5869930982589722,
-0.6517999172210693,
0.211549714207649... |
Detailed Instructions: In this task, you are given a sentence or phrase in English. You must translate it to Xhosa in a way that is equivalent in terms of meaning and grammatically correct.
See one example below:
Problem: Rope and its Usage
Solution: Intambo nomsebenzi ewenzayo.
Explanation: 'Intambo' means rope, 'nomsebenzi ewenzayo' roughly translates to 'the work it does', which is a native way of phrasing it.
Problem: It can be the cause of parting without warning.
Solution: | Ingayintsusa yomahluko ngaphandle kwesilumkiso. | 4 | NIv2 | task872_opus_xhosanavy_translation_eng_xhosa | fs_opt | [
0.5550195574760437,
0.9081926345825195,
0.20910422503948212,
-0.04724797606468201,
-0.5505654215812683,
-0.8858351707458496,
0.8821139931678772,
0.30505168437957764,
-0.5901187062263489,
-0.18652823567390442,
-0.3349519968032837,
-0.0740860253572464,
-0.6311771869659424,
-0.013378109782934... |
In this task, you are given a hypothesis and an update. The hypothesis sentence is a statement that speaks of a socially normative behavior. In other words, it is a generalizing statement about how we expect people to behave in society. The update provides additional contexts about the situation that might UNDERMINE or SUPPORT the generalization. An undermining context provides a situation that weakens the hypothesis. A supporting context provides a situation that strengthens the generalization. Your task is to output 'strengthener' or 'weakener' if the update supports or undermines the hypothesis, respectively
Q: Hypothesis: It's rude to abandon plans you've made with friends about your living situation.
Update: your situations abruptly changes and you can't pursue the plans
A: | weakener | 4 | NIv2 | task937_defeasible_nli_social_classification | zs_opt | [
-0.16992776095867157,
0.7261865139007568,
0.07624386250972748,
-0.06366372853517532,
-0.16778135299682617,
-0.8629438281059265,
0.8978259563446045,
1.1618947982788086,
0.6268774271011353,
-0.31646913290023804,
-1.395681381225586,
-0.26375120878219604,
-0.8636823296546936,
-0.01539360266178... |
You are given a conversation between two people. 'Person1:' and 'Person2:' are used to separate their respective dialogues. Your task is to label each of the continuous conversations done by one speaker with the emotion detected in that particular part. Your labels can be one among the following: 'No emotion', 'surprise', 'happiness', 'sadness', 'anger', 'disgust'.
[Q]: Person1: Tom , you're my good friend , right ?
Person2: Sure . What's the matter with you ?
Person1: The exam ! The terrible exam ! I'm not ready yet . Can you tell me how to get good marks ?
Person2: Don't be so nervous ! Relax ! First , sleep early at night . Second , study two hours a day after school . Third , don't tire yourself out , but don't be lazy . Will that help ?
Person1: Yes , thank you .
[A]: No emotion,No emotion,fear,No emotion,No emotion
[Q]: Person1: Why are you laughing ?
Person2: I can't help it .
[A]: No emotion,No emotion
[Q]: Person1: You're spending a very long time on my hair .
Person2: I'm almost done .
Person1: I needed you to be done and finished an hour ago .
Person2: I'm all done now .
Person1: Are you sure you're finished ?
Person2: I don't understand .
Person1: You did a horrible job on my hair .
Person2: What's wrong with it ?
Person1: I don't like anything about it .
Person2: I apologize .
Person1: That's fine , but I'm not paying for it .
Person2: Excuse me ? Let's try to work this out , okay ?
[A]: | anger,No emotion,No emotion,No emotion,anger,No emotion,anger,No emotion,disgust,No emotion,No emotion,No emotion
| 5 | NIv2 | task1532_daily_dialog_emotion_classification | fs_opt | [
0.2766333222389221,
0.18198326230049133,
-0.050498705357313156,
-0.6461029052734375,
-0.0673791766166687,
-0.8505306243896484,
0.7251901626586914,
0.4513911008834839,
-0.38150084018707275,
-0.20646613836288452,
-0.014351116493344307,
-0.06022258102893829,
-0.33947014808654785,
-0.133418679... |
Detailed Instructions: Given a phrase describing the relationship between two words, extract the words and the lexical relationship between them. The relation has to be of the type 'MemberOf', 'MadeOf', 'Synonym', 'Entails', 'HasA', 'HasProperty', 'PartOf', 'Antonym' or 'IsA'. The output should have the format: word1 relation word2.
Q: flower is a kind of gift
A: | flower IsA gift | 9 | NIv2 | task1510_evalution_relation_extraction | zs_opt | [
-0.22756269574165344,
0.9738955497741699,
0.11659427732229233,
-0.40596145391464233,
-0.24869239330291748,
-1.008617639541626,
0.7216590642929077,
0.2757498025894165,
0.500109076499939,
-0.7881627678871155,
-0.617976188659668,
0.5187432765960693,
-0.8613239526748657,
0.8740633726119995,
... |
Q: In this task, you are given a hateful post in Bengali that expresses hate or encourages violence towards a person or a group based on the protected characteristics such as race, religion, sex, and sexual orientation. You are expected to classify the post into two classes: personal or non-personal depending on the topic.
সৌদি রাজপরিবার হলো ইয়াজিদের বংশধর। এরা আমেরিকা ও ইসরাইলের দালাল।
A: | non-personal | 7 | NIv2 | task1490_bengali_personal_hate_speech_binary_classification | zs_opt | [
-1.177963376045227,
0.4114726483821869,
0.3596096634864807,
0.30424174666404724,
-0.49782073497772217,
0.11490504443645477,
0.9461091756820679,
0.4305056929588318,
-0.21757885813713074,
-0.29679059982299805,
-0.29249972105026245,
0.8415470719337463,
-0.5483007431030273,
-0.3164055347442627... |
You are given a sentence in Hebrew. Your job is to translate the Hebrew sentence into Galician.
[EX Q]: תודה לך ביל גייטס ותודה לך אמא.
[EX A]: Grazas Bill Gates e grazas mamá.
[EX Q]: היחידה הקטנה הזו כאן — זוכרים את השקע הקטן של 12 וולט במכונית שלכם, שטוענת את מה שלא יהיה — גיימבוי, טלפון?
[EX A]: Esta pequena unidade, recordan o pequeno enchufe de 12 V do seu coche, que carga calquera cousa? A Game Boy, o teléfono...
[EX Q]: חיידקים אחרים, על מנת להתגונן, מפתחים הגנות כנגד אותה מתקפה כימית.
[EX A]: | Outras bacterias, para protexerse, evolucionan creando defensas contra ese ataque químico.
| 6 | NIv2 | task1110_ted_translation_he_gl | fs_opt | [
-1.0617680549621582,
0.526025116443634,
-0.6474235653877258,
0.11890936642885208,
-1.100399136543274,
-2.011258602142334,
0.6693020462989807,
0.4573867917060852,
0.4998725950717926,
-0.37748995423316956,
-0.18195949494838715,
1.0289565324783325,
-0.657482922077179,
-0.35637611150741577,
... |
Teacher:You are given a sentence in English. Your job is to translate the English sentence into Persian.
Teacher: Now, understand the problem? Solve this instance: So typically, they want to transmit wealth to the next generation, to their children, or sometimes they want to accumulate wealth because of the prestige, the power that goes with wealth.
Student: | قالبا افراد مایل به انتقال ثروت به نسل بعدی و فرزاندشان هستند ، و دلیل دیگر برای انباشت ثروت به واسطه وجهه و قدرتی است که ثروت به ارمغان می آورد. | 6 | NIv2 | task1093_ted_translation_en_fa | zs_opt | [
-1.2533178329467773,
0.5811183452606201,
0.1798660010099411,
-0.837329626083374,
-0.6627181768417358,
-0.08729282766580582,
0.5191398859024048,
1.0865850448608398,
0.5690011978149414,
-0.41221725940704346,
0.24835193157196045,
0.6775641441345215,
-1.101993441581726,
-0.329356849193573,
0... |
In this task you will be given a claim and a perspective. You should determine whether that perspective supports or undermines the claim. If the perspective could possibly convince someone with different view, it is supporting, otherwise it is undermining.
Q: claim: The Guantanamo Bay detention centre should be closed.
perspective: Guantanamo Bay’s Lack of Transparency Counters American Interests.
A: | support | 4 | NIv2 | task738_perspectrum_classification | zs_opt | [
0.17513152956962585,
0.13814789056777954,
0.28247934579849243,
-0.5347191095352173,
-0.33918026089668274,
-0.6374514698982239,
0.43983638286590576,
0.5040427446365356,
-0.056475117802619934,
0.18511487543582916,
0.11438232660293579,
-0.3570299744606018,
-0.8084244132041931,
0.4439095258712... |
You will be given a definition of a task first, then some input of the task.
Given an English language product review, determine if it is a Good Review or a Bad Review. A good review is one where the product's review has positive tone and Bad review is one where the tone of product's review is negative.
This is the perfect size & weight for making soup or gravies. It heats evenly, has a good grip handle, looks beautiful in my kitchen :) Love it.
Output: | Good Review | 1 | NIv2 | task929_products_reviews_classification | zs_opt | [
-0.5957913994789124,
0.11481915414333344,
0.12946899235248566,
-0.012665756046772003,
0.18290942907333374,
-1.072901964187622,
0.7090303897857666,
0.7193756699562073,
0.2945561408996582,
0.3924300968647003,
-0.009391967207193375,
0.3874199390411377,
-0.27414000034332275,
-0.329716145992279... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.