source
stringlengths 17
501k
| target
stringlengths 0
8.23k
| task_name
stringlengths 8
85
| task_source
stringclasses 4
values | template_type
stringclasses 4
values | template_idx
int64 0
17
| split
stringclasses 3
values |
|---|---|---|---|---|---|---|
You will be given a definition of a task first, then an example. Follow the example to solve a new instance of the task.
Given a sentence in Japanese, provide an equivalent paraphrased version from the original that retains the same meaning.
1975 - 76幎ã®NBAã·ãŒãºã³ã¯ãå
šç±³ãã¹ã±ããããŒã«åäŒã®30çªç®ã®ã·ãŒãºã³ã§ããã
Solution: 1975 - 76幎ã®å
šç±³ãã¹ã±ããããŒã«åäŒã®ã·ãŒãºã³ã¯ãNBAã®ç¬¬30ã·ãŒãºã³ã§ããã
Why? The paraphrase of the original sentence in Japanese is correct and accurate because even though sentence order has been changed, the denotion that the 30th season of the NBA occurred in the years 1975 - 1976 is retained.
New input: ã¢ã«ãŒã³ãã³æ¿åºã¯æŠäºã®æåŸã®æ¥ãŸã§äžç«ãä¿ã£ããã1945幎ã«ãã€ãããã«ã®ãŒããããŠãŽã£ã·ãŒãã©ã³ã¹ããéãåºããããã®æå°è
ãã¡ã®åå
¥ãéãã«å®¹èªããã
Solution:
|
ã¢ã«ãŒã³ãã³æ¿åºã¯æŠäºã®æåŸã®æ¥ãŸã§äžç«ãä¿ã£ããã1945幎ã«ãŽã£ã·ãŒããã©ã³ã¹ããã«ã®ãŒããããŠãã€ãããéããããã®æå°è
ãã¡ã®åå
¥ãé»ã£ãŠå®¹èªããã
|
task776_pawsx_japanese_text_modification
|
NIv2
|
fs_opt
| 0
|
train
|
Detailed Instructions: Given a sentence in Japanese, provide an equivalent paraphrased version from the original that retains the same meaning.
See one example below:
Problem: 1975 - 76幎ã®NBAã·ãŒãºã³ã¯ãå
šç±³ãã¹ã±ããããŒã«åäŒã®30çªç®ã®ã·ãŒãºã³ã§ããã
Solution: 1975 - 76幎ã®å
šç±³ãã¹ã±ããããŒã«åäŒã®ã·ãŒãºã³ã¯ãNBAã®ç¬¬30ã·ãŒãºã³ã§ããã
Explanation: The paraphrase of the original sentence in Japanese is correct and accurate because even though sentence order has been changed, the denotion that the 30th season of the NBA occurred in the years 1975 - 1976 is retained.
Problem: ãããä»ã·ãŒãºã³æåŸã®AFLãã£ã³ããªã³ã·ããã§ãããæåã®ã¹ãŒããŒããŠã«ã¯1966幎ã·ãŒãºã³ã«ç¶ããŸããã
Solution:
|
ããã¯ã·ãŒãºã³ãå®äºããæåŸã®AFLãã£ã³ããªã³ã·ããã§ããã1966幎ã·ãŒãºã³ã«ç¶ããŠæåã®ã¹ãŒããŒããŠã«ã§ããã
|
task776_pawsx_japanese_text_modification
|
NIv2
|
fs_opt
| 4
|
train
|
Detailed Instructions: Given a sentence in Japanese, provide an equivalent paraphrased version from the original that retains the same meaning.
Q: ç»æçãªãã¹ããã·ã¥ãïŒ1994幎ïŒã®4æ²ç®ãš3æ²ç®ã®ã·ã³ã°ã«ã§ãã
A:
|
ç»æçãªã¢ã«ãã ãSmashãïŒ1994幎ïŒã®4æ²ç®ãš3æ²ç®ã®ã·ã³ã°ã«ã§ãã
|
task776_pawsx_japanese_text_modification
|
NIv2
|
zs_opt
| 9
|
train
|
Detailed Instructions: Given a sentence in Japanese, provide an equivalent paraphrased version from the original that retains the same meaning.
Q: ã¹ãºã¡ã¯ã¿ãŒããŒã«ãšãªã¶ãã¹ãèžãèŠã€ããããã«éæ³ã®çŸ
éç€ã§è§£æŸãããããšãã§ãããšç¢ºä¿¡ãããŸãã
A:
|
ã¹ãºã¡ã¯ã¿ãŒããŒã«ããšãªã¶ãã¹ãèžãèŠã€ããããã«éæ³ã®çŸ
éç€ã䜿ãããšã«ãã£ãŠè§£æŸãããããšãã§ãããšç¢ºä¿¡ãããŸãã
|
task776_pawsx_japanese_text_modification
|
NIv2
|
zs_opt
| 9
|
train
|
You will be given a definition of a task first, then an example. Follow the example to solve a new instance of the task.
Given a sentence in Japanese, provide an equivalent paraphrased version from the original that retains the same meaning.
1975 - 76幎ã®NBAã·ãŒãºã³ã¯ãå
šç±³ãã¹ã±ããããŒã«åäŒã®30çªç®ã®ã·ãŒãºã³ã§ããã
Solution: 1975 - 76幎ã®å
šç±³ãã¹ã±ããããŒã«åäŒã®ã·ãŒãºã³ã¯ãNBAã®ç¬¬30ã·ãŒãºã³ã§ããã
Why? The paraphrase of the original sentence in Japanese is correct and accurate because even though sentence order has been changed, the denotion that the 30th season of the NBA occurred in the years 1975 - 1976 is retained.
New input: 圌ã¯1685幎ã«ã±ããã¯å·ã«åŒã£è¶ãããã°ããã®éãã¥ãŒãã©ã³ã¹ã«äœãã§ããŸããã
Solution:
|
圌ã¯1685幎é ã«ã±ããã¯ã«åŒã£è¶ãããã°ããã®éãã¥ãŒãã©ã³ã¹ã«äœãã§ããŸããã
|
task776_pawsx_japanese_text_modification
|
NIv2
|
fs_opt
| 0
|
train
|
Given the task definition and input, reply with output. Given a sentence in Japanese, provide an equivalent paraphrased version from the original that retains the same meaning.
以åã®ç·šéè
ã«ã¯ãMichel FoucaultãEmmanuel Le Roy LadurieãGeorges DumezilãFrancois JacobãJacques Le GoffãFrancois FuretãRaymond AronããããŸãã
|
éå»ã®ç·šéè
ã«ã¯ãRaymond AronãGeorges DumezilãFrancois JacobãMichel FoucaultãEmanuel Roy LadurieãFrancois Furetãããã³Jacques Le Goffãå«ãŸããŸãã
|
task776_pawsx_japanese_text_modification
|
NIv2
|
zs_opt
| 5
|
train
|
Given a sentence in Japanese, provide an equivalent paraphrased version from the original that retains the same meaning.
One example: 1975 - 76幎ã®NBAã·ãŒãºã³ã¯ãå
šç±³ãã¹ã±ããããŒã«åäŒã®30çªç®ã®ã·ãŒãºã³ã§ããã
Solution is here: 1975 - 76幎ã®å
šç±³ãã¹ã±ããããŒã«åäŒã®ã·ãŒãºã³ã¯ãNBAã®ç¬¬30ã·ãŒãºã³ã§ããã
Explanation: The paraphrase of the original sentence in Japanese is correct and accurate because even though sentence order has been changed, the denotion that the 30th season of the NBA occurred in the years 1975 - 1976 is retained.
Now, solve this: 209幎ã®åãã«ãSun Quan Sunã¯ãQuanã®åйLady Sunãšçµå©ãã圌ãšLiu Beiãšã®åçé¢ä¿ã匷åããŸããã
Solution:
|
209幎ã®åãã«ããµã³ã«ã³ã¯ããµã³ã«ã³ã®åйã¬ãã£ãŒãµã³ãšçµå©ãã圌ãšååãšã®åçé¢ä¿ã匷åããŸããã
|
task776_pawsx_japanese_text_modification
|
NIv2
|
fs_opt
| 6
|
train
|
Given a sentence in Japanese, provide an equivalent paraphrased version from the original that retains the same meaning.
Example Input: DantÚsã®å人ãFernand MondegoïŒSidney BlackmerïŒã¯åœŒãååæã«é£ããŠè¡ããŸãã
Example Output: DantÚsã®å人Sidney BlackmerïŒFernand MondegoïŒã¯åœŒãååæã«é£ããŠè¡ããŸãã
Example Input: Taizhou StationïŒæµæ±çïŒã¯äžè¯äººæ°å
±ååœæµæ±çTaizhouã«ããYongtaiwen Railwayã®ééé§
ã§ãã
Example Output: å°å·é§
ïŒæµæ±çïŒã¯ãäžè¯äººæ°å
±ååœãæµæ±çå°å·åžã«äœçœ®ããæ°žæ³°ç·ã®ééé§
ã§ãã
Example Input: ãééºäŒæ§çç¶èµ€è¡çå¢å çããšããçšèªããææã§ã¯ãããŸãã䜿çšãããããšã¯ãã£ãã«ãããŸããã
Example Output:
|
ãééºäŒæ§çç¶èµ€è¡ççããšããçšèªã¯ãææã§ã¯ãããŸããããã£ãã«äœ¿çšãããŸããã
|
task776_pawsx_japanese_text_modification
|
NIv2
|
fs_opt
| 3
|
train
|
Instructions: Given a sentence in Japanese, provide an equivalent paraphrased version from the original that retains the same meaning.
Input: ãããŒããŠããŒã¯2004幎ã«ããã¥ãŒããã¬ã€ãã§ã«ãã©ããã«ãã£ãŠæãããæ¶ç©ºã® "ãµã¿ããŒãã€ãã©ã€ã"ã®ãã£ã©ã¯ã¿ãŒã®ååã§ãã
Output:
|
ãããŒããŠããŒã¯2004幎ã«ããã¥ãŒããã¬ã€ãã§ã«ãã©ããã«ãã£ãŠæãããæ¶ç©ºã®ããµã¿ããŒãã€ãã©ã€ããã®ãã£ã©ã¯ã¿ãŒã®ååã§ãã
|
task776_pawsx_japanese_text_modification
|
NIv2
|
zs_opt
| 3
|
test
|
Given a sentence in Japanese, provide an equivalent paraphrased version from the original that retains the same meaning.
ã¯ã£ã³ãã¥ã¹Metellusã¯ããŒãã®æ¿æ²»å®¶ãšäžè¬ã®ã«ããŠã¹Metellusã¯ã©ãŠã³ã®æ¬¡ç·ã§ããã
|
ã¯ã£ã³ãã¥ã¹Metellusã¯ãããŒãã®æ¿æ²»å®¶ã®æ¬¡ç·ãäžè¬ã«ããŠã¹Metelluså ã§ããã
|
task776_pawsx_japanese_text_modification
|
NIv2
|
zs_opt
| 0
|
validation
|
In this task, you are given an utterance, which is a part of a conversation between a user and an agent. Your job is to detect the speaker. The user usually commands the agent to schedule events, retrieve event information, and check the schedule. While the agent's response is an answer to the user's questions or follow-up questions on the user's command. Answer with "User" or "Agent".
[EX Q]: This meeting time is okay for me
[EX A]: user
[EX Q]: Hi can you help me create an event
[EX A]: user
[EX Q]: Put on my calendar an item called Sell Stuff Online starting on Friday at 9 am and lasting for 6 hours, and mark me as out of office during that time
[EX A]:
|
user
|
task1599_smcalflow_classification
|
NIv2
|
fs_opt
| 6
|
train
|
Instructions: In this task, you are given an utterance, which is a part of a conversation between a user and an agent. Your job is to detect the speaker. The user usually commands the agent to schedule events, retrieve event information, and check the schedule. While the agent's response is an answer to the user's questions or follow-up questions on the user's command. Answer with "User" or "Agent".
Input: I need a room at NERD for the interviews
Output:
|
user
|
task1599_smcalflow_classification
|
NIv2
|
zs_opt
| 3
|
train
|
Instructions: In this task, you are given an utterance, which is a part of a conversation between a user and an agent. Your job is to detect the speaker. The user usually commands the agent to schedule events, retrieve event information, and check the schedule. While the agent's response is an answer to the user's questions or follow-up questions on the user's command. Answer with "User" or "Agent".
Input: I've deleted your event matching "lunch" on December 1st.
Output:
|
agent
|
task1599_smcalflow_classification
|
NIv2
|
zs_opt
| 3
|
train
|
Instructions: In this task, you are given an utterance, which is a part of a conversation between a user and an agent. Your job is to detect the speaker. The user usually commands the agent to schedule events, retrieve event information, and check the schedule. While the agent's response is an answer to the user's questions or follow-up questions on the user's command. Answer with "User" or "Agent".
Input: I need to call Clark Kent.
Output:
|
user
|
task1599_smcalflow_classification
|
NIv2
|
zs_opt
| 3
|
train
|
In this task, you are given an utterance, which is a part of a conversation between a user and an agent. Your job is to detect the speaker. The user usually commands the agent to schedule events, retrieve event information, and check the schedule. While the agent's response is an answer to the user's questions or follow-up questions on the user's command. Answer with "User" or "Agent".
Example Input: May 8,Wed 1:00 PM - 1:30 PM
Example Output: user
Example Input: Add golf on Saturday fro 10:00 AM to 3:00 PM.
Example Output: user
Example Input: Who are Jake's reports
Example Output:
|
user
|
task1599_smcalflow_classification
|
NIv2
|
fs_opt
| 3
|
train
|
In this task, you are given an utterance, which is a part of a conversation between a user and an agent. Your job is to detect the speaker. The user usually commands the agent to schedule events, retrieve event information, and check the schedule. While the agent's response is an answer to the user's questions or follow-up questions on the user's command. Answer with "User" or "Agent".
Ok thank you
|
user
|
task1599_smcalflow_classification
|
NIv2
|
zs_opt
| 0
|
train
|
Given the task definition and input, reply with output. In this task, you are given an utterance, which is a part of a conversation between a user and an agent. Your job is to detect the speaker. The user usually commands the agent to schedule events, retrieve event information, and check the schedule. While the agent's response is an answer to the user's questions or follow-up questions on the user's command. Answer with "User" or "Agent".
Decline the invite to vacation in Italy
|
user
|
task1599_smcalflow_classification
|
NIv2
|
zs_opt
| 5
|
train
|
Teacher:In this task, you are given an utterance, which is a part of a conversation between a user and an agent. Your job is to detect the speaker. The user usually commands the agent to schedule events, retrieve event information, and check the schedule. While the agent's response is an answer to the user's questions or follow-up questions on the user's command. Answer with "User" or "Agent".
Teacher: Now, understand the problem? Solve this instance: Good now change it to one hour
Student:
|
user
|
task1599_smcalflow_classification
|
NIv2
|
zs_opt
| 6
|
train
|
Instructions: In this task, you are given an utterance, which is a part of a conversation between a user and an agent. Your job is to detect the speaker. The user usually commands the agent to schedule events, retrieve event information, and check the schedule. While the agent's response is an answer to the user's questions or follow-up questions on the user's command. Answer with "User" or "Agent".
Input: The time is currently 1:24.
Output:
|
agent
|
task1599_smcalflow_classification
|
NIv2
|
zs_opt
| 3
|
test
|
In this task, you are given an utterance, which is a part of a conversation between a user and an agent. Your job is to detect the speaker. The user usually commands the agent to schedule events, retrieve event information, and check the schedule. While the agent's response is an answer to the user's questions or follow-up questions on the user's command. Answer with "User" or "Agent".
Here are some times when you and Carl are free.
agent
At the end of my day on Sunday schedule Cook for the Week.
user
You haven't helped me with anything yet!
|
user
|
task1599_smcalflow_classification
|
NIv2
|
fs_opt
| 0
|
validation
|
TASK DEFINITION: In this task, you are given dvd product reviews in Japanese language. The goal is to classify the review as "POS" if the overall sentiment of the review is positive or as "NEG" if the overall sentiment of the review is negative.
PROBLEM: äºèŠã§ããªãããŒããŒ&è¡æçãªã·ãŒã³ . ãã®æ ç»ã«ã¯2ã€ã®ã·ã§ããã³ã°ãªèŠãå Žã®ã·ãŒã³ããããä»ã¯ããã«è³ãå眮ããšåŸæ¥è«ã«ããæããŸããäžæ¹ã®èŠãå Žã«ã€ããŠã¯æ¥æ¬ã®ã矩çµèšãã®åŒæ
¶ã®ææãã€ã¡ãŒãžããã®ããªãšæããŸãããããåãããŸãããäžåœã®å²å®ã«ãã話ãããã¹ããŒã¢ãŒã·ã§ã³ã®æ åãšä¿³åªã®è¡šæ
ãåããåãã§ããæ²ãããã²ãããçŸããã§ããäŸãæ¥æ¬ã®è©±ããªãªãžãã«ã§ãã£ããšããŠããããã¡ãã®æ¹ãäŒèª¬åããŠããããã®äœã§ãã2ã€ã®èŠãå Žã¯æ±ºããŠäžæç©è¶£å³ã®ãã®ã§ã¯ãããŸãããæšæž¬ã§ãããæ ç»ã®äœãæã2人ã®ä¿³åªã®çŽç²ãªçŸããããã£ããããã§ æ°žé ã«ãããã£ãã®ã§ã¯ãªãã§ãããããã¢ã¯ã·ã§ã³æ ç»ãšããŠã¯ã¹ããŒãªãŒã«ç¡é§ãªæãç·©æ
¢ãªæããªããéå±ããªãå±éã§ãããã ãã¢ã¯ã·ã§ã³æ ç»ç¬ç¹ã®ãéœå䞻矩çãªãšããããªããããŒããŒãå±éºãå¯ç¥ãçŽ æ©ãåããã¿ãããããããŸãããããããããããªãäºèŠã§ããªã?ããšäžæè°ãªãããã®ãšããããããŸããããªãŠããçãªåçŽã«æ¥œãããçå¿«ã¢ã¯ã·ã§ã³ã芳ãã人ãªãããã¯éããšæããŸãã
SOLUTION: POS
PROBLEM: ãã¬ããèŠããªããªã£ãŠããŸããŸãã . ãã®DVDã¯ãšã£ãŠãããã§ããéé·ãšç·ã®åãšå¥³ã®åã®äŒè©±ã¯ãŸã çè§£ããŠããªãã®ã§ããïŒïŒæ³åãªã®ã§ïŒæ²¹å§ã·ã§ãã«ããã€ãŒã«ããŒããåºãŠãããšæãã·ã§ãã«ã®åœ¢ã«ããŠãã~ã£ãšãã£ãŠããŸãïŒç¬ïŒ ãããèŠããŠããéã«æŽæ¿¯ãæé€ãªã©ã®å®¶äºãã§ããããã«ãªããŸããã ã§ããæ®éã®ãã¬ãçªçµã«ãããšæã£ãŠããŸããéåžžã®æŸéã¯ã»ãšãã©èŠããªããªã£ãŠããŸããŸããã»ã»ã»ã»
SOLUTION: POS
PROBLEM: åãªãªãŒã¹(2å±€ãVC-1 DTS-HD5.1ch) . HD-DVDã§ãæ¢ã«å®æåºŠã®é«ãã£ãæ¬äœã åå Žçã»23å远å ããããšã¯ã¹ãã³ãããçãäž¡åé²ã ååã®ã¯ã©ã·ãã¯ãªé°å²æ°ãããCGãé§äœ¿ããæ åãŸã§èŠæã¯æºèŒã Blu-rayåã«ããããã¹ã¬ã¹é³å£°åé²ããæŽã«ç£šããããã£ãå°è±¡ã 枯ããåºèªåŸã®åææ åã®èæ¯ã®è«§èª¿ãåäžã é«é«å³¶äžéžåŸãçãèŽå°ã«åãããã¢ã³ã®ããšã«ãéãç«ã¡èŸŒããæ£®ã®äžããã³ã³ã°ãç»å Žããã·ãŒã³ã¯èŠç©ã ãã®å³¶ã®äžã§ã®äžå€§åéºã·ãŒã³ã¯ãèŠäºãªç«äœæãšè§£åæãå³ãããã 被åäœã«ç
§ãç
§æã®ç確ããšçžãŸã£ãŠãæææ¬åœã«è§Šããããªçšã®ãªã¢ã«ãªé«ç»è³ªã 象城çãªãšã³ãã€ã¢ã»ã¹ããŒããã«ã§ã®æŠéã·ãŒã³ãèŠæã§ããããå
ã®é«é«å³¶ã§ã®ã·ãŒã³ãäœãã«çŽ æŽãããçºã«ãå€å°èŠå£ãããã æ°ãã«DTS-HD5.1ïœïœã§åé²ããããµãŠã³ãã¯ããã€ãããã¯ã è¶³é³ãªã©éäœé³ãå°é¿ãã®ããã«é¿ãã ã©ã¹ãã®æŠéæ©ãã³ã³ã°ãæ»æããå Žé¢ã§ã¯ããµã©ãŠã³ã广ãé«ã楜ãããã å
šäœçã«ã¯ç¹çްãã«ã¯æ¬ ããå°è±¡ã®ãµãŠã³ãèšèšã§ã¯ããããæŽŸæãªäœé³ã¯å€§ãã«å³ãããã ç¹å
žã¯ãU-ã³ã³ãããŒã«(Picture in Picture)ãªã©ããã¡ããšæŒãããŠãããå°ãªãã ãã ãããã ãé·å°ºã§2ããŒãžã§ã³ãé«ç»è³ªã§åããŠãªããšä»æ¹ããªãã
SOLUTION:
|
POS
|
task486_cls_japanese_dvd_classification
|
NIv2
|
fs_opt
| 8
|
train
|
Given the task definition, example input & output, solve the new input case.
In this task, you are given dvd product reviews in Japanese language. The goal is to classify the review as "POS" if the overall sentiment of the review is positive or as "NEG" if the overall sentiment of the review is negative.
Example: ïŒ~ïŒè©±ã§ã¯ãŸã£ãŠããŸãããšã§ãããã . éã®ïŒå·ã®ãã£ã©ã¯ã¿ãŒãã¶ã€ã³ã奜ããªäººã¯ããã«ã¯ãŸã£ãŠããŸãäœåã ãšæããŸãããã®ä»ã®äººã§ãäžåºŠèгããã¯ãŸã£ãŠããŸããšæããŸããïŒ~ïŒè©±ããåé²ãããŠããªãã®ãå®ã«æ®å¿µã§ããã話ã®å
容ãäžç芳ã¯ããã«ã€ããããšæããŸãããŽã¡ã³ã·ãããªã©ã®ã¡ã«ããã¯ãã¶ã€ã³ãå®ã«ããã§ããŠããŸãããã³ããããŠååãæã«ãããŸããããã¡ãªã¿ã«ç§ã¯ã¢ã¬ãã¯ã¹ã»ããŠãä»ã®ãšããäžçªæ°ã«å
¥ã£ãŠãããã£ã©ã¯ã¿ãŒã§ãã
Output: POS
The overall sentiment of the review is positive as the reviewer refers to the dvd with positive expressions, such as 'ïŒ~ïŒè©±ã§ã¯ãŸã£ãŠããŸãããšã§ãããã', 'ããã§ããŠããŸã' etc. Hence, the label is 'POS'.
New input case for you: ç£ç£ãæªããšããããã¯æ§æãæªã . ããŒãäœç»ã鳿¥œã¯ãã¯ããžããªãã©ã³ãã§ãããè¯ãã§ããé³å£°ãè³åŠäž¡è«ã§ãããç§çã«ã¯ãããã蚱容ã§ããç¯å²ã§ããããããã®ããçšåºŠã®ã¯ã©ãªãã£ãæã£ãäœåãé§äœãããããŠãã®ã¯æ§æã§ããèµ·æ¿è»¢çµããŸã£ããã§ããŠããªãã§ããåäœãç¥ããªãç§ãªã©ã¯ç¹ã«ãã¢ã¬ã³ãç¶èŠªã«ãã£ãããšãžã®åæ©ããã«ãŒã®æ£äœã«ã€ããŠã®èæ¯ããŸã£ããããããªããçµå±ã²ãã®ç®çã¯ãªããªã®ãïŒçãèãããã«æ
ããŠãã®ïŒïœïŒããªã©äžåã¿ãã ãã§ã¯æ£çŽãªã«ããªã«ããããããªããŸãŸçµãã£ãŠããŸããŸããïŒ2床ç®ãèŠãããšã¯ããããŸãããã»ã»ã»ïŒ ãæ§æã¯ã ããæ
åœããã®ãã¯ç¥ããªãã®ã§ãªããšããããŸããããæçµçã«ã¯ç£ç£ãOKãã ããŠãããã»ã»ã»ãšããããŸãã®ã§ãç£ç£ã®è²¬ä»»ãšããããšã«ãªããŸããèãã°åäœã¯äžçïŒå€§ãã¡ã³ã¿ãžãŒãšãŸã§ããããŠããããã§ããããã®ã¢ãã¡ããã¯åäœã«ããããŠãªãã®èå³ã沞ããŠããŸããã話ããŸãšããããªãã®ãªãã°ïŒéšæ§æã«ãããªããªããªãã§ãã¡ããšäœãããã§ã»ããã£ãã§ããç®å
ã®éç®åœãŠã®äœåãšããããããããªãããžããªã®è©äŸ¡ãèããäžããŠãã®ã§ã¯ãªãããšæããŸãã
Output:
|
NEG
|
task486_cls_japanese_dvd_classification
|
NIv2
|
fs_opt
| 1
|
train
|
Teacher:In this task, you are given dvd product reviews in Japanese language. The goal is to classify the review as "POS" if the overall sentiment of the review is positive or as "NEG" if the overall sentiment of the review is negative.
Teacher: Now, understand the problem? Solve this instance: ãªãŒïŒé§äœ . ã€ãŸããªãäœåã§ããèŠã䟡å€ããªãããäœãããã®ã ãåãããªããã ã®é§äœã§ãããã€ã»ãžã§ã³ãžã§ã¯ãã ã§ãããã©ãã以å€ã®ãšããã¯èŠæãªãã
Student:
|
NEG
|
task486_cls_japanese_dvd_classification
|
NIv2
|
zs_opt
| 6
|
train
|
In this task, you are given dvd product reviews in Japanese language. The goal is to classify the review as "POS" if the overall sentiment of the review is positive or as "NEG" if the overall sentiment of the review is negative.
ã¹ã«ããšããªãæ ç»ã§ãã . 芳ãŠãã人ã飜ããããªãæ ç»ã§ããããç©äºãäœã解決ããªããŸãŸã«çµäºããŠããŸããããªã©ã¹ãã§ã¹ã«ããšããŸããã§ããïŒãããç£ç£ã®çããªã®ã§ããããïŒã
|
NEG
|
task486_cls_japanese_dvd_classification
|
NIv2
|
zs_opt
| 0
|
train
|
Detailed Instructions: In this task, you are given dvd product reviews in Japanese language. The goal is to classify the review as "POS" if the overall sentiment of the review is positive or as "NEG" if the overall sentiment of the review is negative.
Q: ãããããïŒ . åéæŠç£ç£ã®æ ç»ã奜ãã ã£ãã®ãšãäºåç·šã®ããžã¥ã¢ã«ã綺éºã ã£ãã®ã§æåŸ
ããŠãã®ã§ãããææ³ã¯ã»ã»ã»ãããããïŒ ç¢ºãã«ãªãŒããã³ã°ã¯ç®ãåºããããªè²åœ©ãšäž»äººå
¬ïŒäººã®ç°åžžãªã·ãã¥ãšãŒã·ã§ã³ã«å§åãããŸããããã®ç·åŒµæãç¶ããªãã ãã®åŸã¯ããããããªãã¿ãªã»ãããšåç¶ãšããã»ã©ãã¿ãªäººéé¢ä¿ã®ãã¿ãã¿ãªãã©ãã«ãªã£ã¡ãã£ãŠãã
A:
|
NEG
|
task486_cls_japanese_dvd_classification
|
NIv2
|
zs_opt
| 9
|
train
|
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 are given dvd product reviews in Japanese language. The goal is to classify the review as "POS" if the overall sentiment of the review is positive or as "NEG" if the overall sentiment of the review is negative.
ïŒ~ïŒè©±ã§ã¯ãŸã£ãŠããŸãããšã§ãããã . éã®ïŒå·ã®ãã£ã©ã¯ã¿ãŒãã¶ã€ã³ã奜ããªäººã¯ããã«ã¯ãŸã£ãŠããŸãäœåã ãšæããŸãããã®ä»ã®äººã§ãäžåºŠèгããã¯ãŸã£ãŠããŸããšæããŸããïŒ~ïŒè©±ããåé²ãããŠããªãã®ãå®ã«æ®å¿µã§ããã話ã®å
容ãäžç芳ã¯ããã«ã€ããããšæããŸãããŽã¡ã³ã·ãããªã©ã®ã¡ã«ããã¯ãã¶ã€ã³ãå®ã«ããã§ããŠããŸãããã³ããããŠååãæã«ãããŸããããã¡ãªã¿ã«ç§ã¯ã¢ã¬ãã¯ã¹ã»ããŠãä»ã®ãšããäžçªæ°ã«å
¥ã£ãŠãããã£ã©ã¯ã¿ãŒã§ãã
Solution: POS
Why? The overall sentiment of the review is positive as the reviewer refers to the dvd with positive expressions, such as 'ïŒ~ïŒè©±ã§ã¯ãŸã£ãŠããŸãããšã§ãããã', 'ããã§ããŠããŸã' etc. Hence, the label is 'POS'.
New input: +Î±ãæ¬²ãããšãã . ç¥èšã§æåãªãã«ãã£ãšãã«ããã³ç¥æ®¿ãåãäžããããŠããã
Solution:
|
POS
|
task486_cls_japanese_dvd_classification
|
NIv2
|
fs_opt
| 0
|
train
|
Given the task definition, example input & output, solve the new input case.
In this task, you are given dvd product reviews in Japanese language. The goal is to classify the review as "POS" if the overall sentiment of the review is positive or as "NEG" if the overall sentiment of the review is negative.
Example: ïŒ~ïŒè©±ã§ã¯ãŸã£ãŠããŸãããšã§ãããã . éã®ïŒå·ã®ãã£ã©ã¯ã¿ãŒãã¶ã€ã³ã奜ããªäººã¯ããã«ã¯ãŸã£ãŠããŸãäœåã ãšæããŸãããã®ä»ã®äººã§ãäžåºŠèгããã¯ãŸã£ãŠããŸããšæããŸããïŒ~ïŒè©±ããåé²ãããŠããªãã®ãå®ã«æ®å¿µã§ããã話ã®å
容ãäžç芳ã¯ããã«ã€ããããšæããŸãããŽã¡ã³ã·ãããªã©ã®ã¡ã«ããã¯ãã¶ã€ã³ãå®ã«ããã§ããŠããŸãããã³ããããŠååãæã«ãããŸããããã¡ãªã¿ã«ç§ã¯ã¢ã¬ãã¯ã¹ã»ããŠãä»ã®ãšããäžçªæ°ã«å
¥ã£ãŠãããã£ã©ã¯ã¿ãŒã§ãã
Output: POS
The overall sentiment of the review is positive as the reviewer refers to the dvd with positive expressions, such as 'ïŒ~ïŒè©±ã§ã¯ãŸã£ãŠããŸãããšã§ãããã', 'ããã§ããŠããŸã' etc. Hence, the label is 'POS'.
New input case for you: ãªã¢ã·ã¹2äœç®ã®ã©ã€ãŽãã㪠. ãªã¢ã·ã¹ã®1æ¬ç®ã®ãããªãLIVE BY THE SEAããèŠãŠå¿åºã©ã€ãŽã®é
·ããã³ãã ãšæããŸããã åœæã2äœç®ã®ã¢ã«ãã ãåäœã ãšæã£ãŠããã®ãäžæ°ã«éããŸãããã ã¢ã«ãã çºè¡šåŸã®ã©ã€ãŽãåãããã®ãããªã¯ã©ã€ãŽã®è奮ãäžæãæããŠããã éã«æãŸããã€ã³ã¿ãã¥ãŒããããæãããã®ã§ã¯ãããŸããã
Output:
|
POS
|
task486_cls_japanese_dvd_classification
|
NIv2
|
fs_opt
| 1
|
train
|
TASK DEFINITION: In this task, you are given dvd product reviews in Japanese language. The goal is to classify the review as "POS" if the overall sentiment of the review is positive or as "NEG" if the overall sentiment of the review is negative.
PROBLEM: ãããããªããšãŸããªãã . ä»äºãå¿ãã人ã¯ãè»œãæ°æã¡ã§èŠãªãã§äžããã å¿
ãå¯äžè¶³ã«ãªããŸãããã~ã£ïŒããã§ãã®äººãïŒïŒã ããŸããã~ãããããã ã£ãã®ã~ãããã~æ»ã¬~殺ãããã ãæ¬¡ã®å·»ã¯ã©ããªãã®~ãã£ãŠæãã§ãããããŸããã ãã³ã¹ãããã§ããããã䞻人å
¬ãšãç»å Žäººç©ãã€é£¯é£ã£ãŠããã€å¯ãŠãã ~ ãããªã«äººéåãããïŒïŒãšã«ããã·ãŒãºã³IIãIII絶察èŠãŠããŸããŸããã
SOLUTION: POS
PROBLEM: ã«ã«ãã»ã ãŒã㌠. ããªãã«ã«ãã§ããïœïœ äºä»¶ããã£ãŠãçç±ããã£ãŠãçŽåŸã§çµããã®ã奜ããªç§ã«ã¯ãçè§£äžèœã£ãœãã§ãã ããããããããŠäºä»¶ããã£ãŠãçç±ããã£ãŠãçŽåŸã§çµãã£ãŠãã®ããã ããã©ã»ããªãŒãåºæŒããŠãŸãã ããã¯ãã€ã³ãâUPã§ãïœïœ
SOLUTION: NEG
PROBLEM: èŠãã«å ªããªã . è°é·ãšèŠé·ã®éã«ãããã£ãé¢ä¿ãããã®ãçŸæ©è²ã£ãŠããããæãå¿
èŠæ§ãæããªããããããè€ç·ã§ç©èªã«éèŠãªäœããããããã«ãæããªãã ã®ã«ããŒãã¯ãçžå€ãããã®èª¬æå£èª¿ããã£ãã誰ã«å¯ŸããŠäœã説æããŠããã®ã ãããããããªãïŒäžæè°ãªãã£ããã ã äžå
µåçžæã«æ¿æããã«ã¬ãªã¯ãäœãŠåšã®å°ããªä»£è¡šãªãã ãããããããå£ããããåŠçŽå§å¡é·ã§ãããïŒç¬ïŒ äžåœã®é·ã§ããã°ãç³ã§äžè¹Žåºæ¥ããããã®ãªãŒã©ãæã£ãŠããªãã®ãïŒ ã¢ã¹ã©ã³ãã¢ã¹ã©ã³ã§ïŒè©±ã®çµããã§ãå ã
ãšèªåã®äºã¢ã¹ã©ã³ã»ã¶ã©ã£ãŠèšã£ãŠããããããã¢ã¬ãã¯ã¹ã£ãŠããåœåã©ããããã ïŒ ã®ã«ããŒãã«ãã¬ããããå¥ã«ããããã®ãïŒïŒå·»ã§ãã«ã¬ãªãäœåºŠãã¢ã¹ã©ã³ã£ãŠé£åŒããŠããã⊠äœãšèšã£ãŠãä»åã®èŠé·ïŒã¿ãªã¢ïŒã®ç¡èœã¶ãã«è±åžœããã æèœã ãšããåè©å€ã ã£ãïŒã®ã«ãïŒè©±ã§ã®ç¡èœèŠé·ã£ã·ãã¯äœãã«ãæããç¡ãã£ãã éåžžèŠé·ãèŠã®çµ¶å¯Ÿæš©ãã¡äœãããåªå
æš©ãããã¯ããªã®ã«ãè°é·ã«äºããæ¯èš±å¯ãããã£ããããã®ã¯å€ã§ãã ãŸããŠããã®éã«ç·å¥³é¢ä¿ïŒïŒè©±ïŒããã£ããšããŠãããã®è¡çºã¯ïŒæµã®å¥³ã«èªãé¥ããŠããã«éããªãã èŠé·ã®åæºã¯èŠå
ã®å£«æ°ã«å€§ããæªã圱é¿ãåºãããè°é·ã®æäººãæããããç©è
°ã¯å¥³ãšããŠãåšãäœãã ç¡èœèŠé·ã©ãããïŒæµã®å¥³ã§ãããªãã
SOLUTION:
|
NEG
|
task486_cls_japanese_dvd_classification
|
NIv2
|
fs_opt
| 8
|
train
|
You will be given a definition of a task first, then some input of the task.
In this task, you are given dvd product reviews in Japanese language. The goal is to classify the review as "POS" if the overall sentiment of the review is positive or as "NEG" if the overall sentiment of the review is negative.
ãã¥ãŒã«ã¬ããã¢ã®çŸããå°å³¶ãã¡ãšå€§æç£ç£ã®çãã®æ å . åç°ç¥äžããã¯80幎代æå€§ã®ããã€ã³ã®ã²ãšãã倧æç£ç£ãå°Ÿé3éšäœã§80幎代ã¯çµ¶é æããã®ãµããã®ãæããããå°å¥³ãã«ã€ãã³ã©ãäœã§ãã ã²ããããããã²ãããçŸãããããå倪平æŽã®æµ·ãå°å³¶ã«äœãçŽ æŽã§ãããããããªã²ãšãã¡ãåŽããã¿ãç çã®ãµã³ã»ããå
æ¯ãæ åãè¶
ãè¶
çŸãããããã ãã§ãçãããŠããŸããŸãã 倧æç£ç£ã®ç§èµã£åã ã£ãã®ã«æããŸãã€ã€åŒéãã髿³è¯äžããã®æŒããçŸå°æ¥ç³»äººãšããæããããå°å¥³ãã®ãšããšãŸã£ããåæ§ã«ã»ã®ããªæ
æ
ãå¯ãåãç¥äžããããŸãã§ãªã€ã³ã«ãŒãã€ã·ã§ã³ã®ããã§ãããµãããšãããŸã¿ãŠããæããè¥è
ã®å¥å
šã§ãããããªèŒãã«ããµããŠããŸãããªãç¥äžããã®ãã©ã 猶ã§ã®å
¥æµŽã·ãŒã³ã¯è©±é¡ã«ãªããŸããã çãã®ã²ãšåç°ç¥äžãšçãã®ç£ç£ã®ã究極ã®ã³ã©ãã¬ã€ãã§ãããçããããã²ãšããã®äœ³åã§ãã
Output:
|
POS
|
task486_cls_japanese_dvd_classification
|
NIv2
|
zs_opt
| 1
|
test
|
You will be given a definition of a task first, then some input of the task.
In this task, you are given dvd product reviews in Japanese language. The goal is to classify the review as "POS" if the overall sentiment of the review is positive or as "NEG" if the overall sentiment of the review is negative.
æïŒå以äžã§ãã . ã¯ã£ããèšã£ãŠããããªã«ãããããªãæ ç»èгãããšããªãã ãªãªãžãã«çã¯èгãããšããªããã芳ãŠã¿ããæ°ã¯ããã ãªãããçµå±æåŸãŸã§ãåå ã¯åãããŸãããã ãïŒ ïŒ¢çŽæ ç»ã§ãããçµããæ¹ãäžéå端âãããâã®ã¯ç®ãã€ãããšããŠãã ãã éãåã£ãŠãã ãã ããããããªãããšãããããããã£ãŠããšãã ãã£ã±ããã£ããæ ç»ã®äž»æšãšããŠããŸã³ããåºãŠããã£ãŠããã ããããªãããšã顿ã«ããŠãããšããŠãã æ®éã®äººéãšããŠãããã§æ®éã¯æ¢ãŸããªãããšãã ããªãã§ãããªäœè£ããŸããŠãã®ããšãããŠãŒã¢ã¢ã®äº€ãæ¹ã倱æã 倱æã倱æã倧倱æãç§ã¯ããããã®æ ç»ã¯æå¥èšããã«èŠããã çµæ§ããããããªãããªæ ç»ã§ãããããšãèŠä»ããŠæ¥œãããã©ã ããã¯ç¡çã§ããããéåºããŠæããã
Output:
|
NEG
|
task486_cls_japanese_dvd_classification
|
NIv2
|
zs_opt
| 1
|
validation
|
Detailed Instructions: 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.
Problem:M: Can I be of any help? ,F: Oh, thank you, Sam. You can be a great help. You know Steve's parents will come to dinner today. This is the first time they have come to visit our new house since our marriage. ,M: Good. What would you like me to do? ,F: Well, I can't make up my mind about what to have for dessert. I made 2 apple pies last night but I tasted one of them this morning and it didn't taste good at all. ,M: You'd better make the pies over again if you don't want Steve's parents to make faces over there desert. ,F: Oh, they wouldn't do that. There are too polite. They would just make me as if nothing was wrong and then complain to Steve about it later. ,M: Look, why don't you just offer some fresh fruit and cheese for dessert? Steve's parents would probably like a healthy dessert.
(A) F: Alright, I'll take them to eat out for the dinner, I don't want they eat the bad taste pineapple pies I made. (B) F: Okay, I'll buy some fruit and vegetables and get well prepared for the lunch. (C) F: I am sorry. You spoke so rapidly that I did not catch the meaning of what you said. Could you repeat that? (D) F: Well, that's a good idea, I'll prepare fresh fruit and cheese for dessert.
Solution:
|
D
|
task611_mutual_multi_turn_dialogue
|
NIv2
|
zs_opt
| 8
|
train
|
Instructions: 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.
Input: M: We still have one more day to go before we return home. What do you want to see tomorrow? ,F: No idea. We've enjoyed both the natural scenery and the city sights? What else is there to see? ,M: How about a factory tour? ,F: You mean visit a factory? ,M: Yes, I heard there's a large car company nearby that is open to the public every weekend. We can go there to see how a car is made and learn more about the car industry. ,F: Good idea. I've always been interested in that. How much is the entrance fee? ,M: I'm not sure, but as far as I know, most factory tours are free. They are sponsored by the car company providing the tour.
(A) F: Wow, we can't miss the opportunity since it is sponsored by the car company. (B) F: Just a minute! I do not quite follow what you are saying, would you mind repeating that? (C) F: Now that it is sponsored by the car company, let's visit it next Monday. (D) F: I don't want to visit it because it is only good for the company and we can learn nothing about the car industry.
Output:
|
A
|
task611_mutual_multi_turn_dialogue
|
NIv2
|
zs_opt
| 3
|
train
|
instruction:
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.
question:
M: Is everything ready for the meeting? ,F: Almost. I'm done with the tables and chairs and I've set up microphones as well. The only thing left to do is put the cups on the table.
(A) M: Hurry up and set the cups! (B) M: You have to set up the tables and chairs now. (C) M: You have put the cups in place. Then it's time to mop the floor. (D) M: Just a minute! I do not quite follow what you are saying, would you mind repeating that?
answer:
A
question:
F: Paul, let's talk about your work experience in South America. What took you there? Was it to improve your Spanish? ,M: Well, I just wanted to find out more about the way people lived there. My spoken Spanish was already pretty good. In fact, I ended up teaching English there.
(A) F: So, your spoken Spanish was not pretty good and you went to South America in order to study English? (B) F: So, you ended up teaching Spanish there and you went to South America to learn about local life? (C) F: Wow, you went to South America to learn about local life? That's great! Let's talk more about your experience! (D) F: Excuse me, could you repeat the question?
answer:
C
question:
M: Hi, there. I think I'm coming down with something. I need some medicine to get me through the next few days. ,F: Wow, you look terrible. What's going on? ,M: Well, my throat hurts. My nose has been running and I have a bad cough. It could just be cold, but I don't know for sure. ,F: You should take some cold medicine and get some rest. You don't look well. ,M: I have a very important job and I can't miss any work. So, give me something strong.
(A) F: I am afraid I did not quite catch what you were saying. Please repeat it. (B) F: Good to know that you'll take a sick leave as you look ill. (C) F: Are you sure you won't rest at home? You really look sick. (D) F: Your throat is perfectly fine? But you sounds unhealthy.
answer:
|
C
|
task611_mutual_multi_turn_dialogue
|
NIv2
|
fs_opt
| 9
|
train
|
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.
Q: F: Have you booked the train tickets? ,M: Yes, the train leaves 4 o'clock and arrives at London at about 10. ,F: What about the plane tickets from London to Paris? ,M: Well, I want your advice. If we take the plane, leaving at midnight, we'll have to pay 260 pounds each. But if we leave at 8 in the morning, we'll have to pay 410. ,F: Can we get to the airport at 11? ,M: Yes, it's only 30 minutes from the railway station to the airport. ,F: Then we'll take the plane, leaving at midnight.
(A) M: We'll have to pay 410 in total for the plane tickets if we we leave at midnight. (B) M: Considering we'd like to avoid unnecessary cost, we'll spend 520 for the plane tickets. (C) M: It must be exhausted to take the plane in the midnight so let's take the morning one, which will cost 820 pounds. (D) M: Iâm sorry, I didnât catch that. Would you mind speaking more slowly?
A:
|
B
|
task611_mutual_multi_turn_dialogue
|
NIv2
|
zs_opt
| 4
|
train
|
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.
One example: M: Excuse me. How much is this suit? ,F: It's on sale today for $750. It's normally $900. ,M: Wow, that is pretty expensive! I was thinking that it might be 4 or 500. ,F: This material is imported from Italy. It's the finest in the world, and if you bought a suit made of this material at many department stores, you would pay about $2000. ,M: Uh-hah. But isn't that the point of coming to a market like this, to get a discount compared to the expensive department stores? Besides I saw a suit just like this one a few stalls down, and they were selling it for $600. I still thought that it was too expensive. ,F: It's possible that the suit you saw was the same color is this one but made of a different material. Unfortunately, our prices are final, and we think there are the lowest anywhere in the city.
(A) M: Okay, I'll buy it at the price you want. Here, it's $600. (B) M: Just a minute! I do not quite follow what you are saying, would you mind repeating that? (C) M: Although the material is imported from France, it is not worth the $750 you want. (D) M: You want $900 for the suit? It's too expensive!
Solution is here: B
Explanation: Options (A) and (B) are not corrct because the suit's price is 900. Option (C) isn't correct because the material was imported from italy.
Now, solve this: M: Hello, reservation desk. How can I help you? ,F: I'd like to book a room. ,M: And what kind of room would you like ma'am? And we got single rooms, double rooms, standard rooms, sweets and Delux sweets. ,F: What's the rate for a standard room? ,M: $79 a night. ,F: I'd like a standard room. ,M: What date would you check in ma'am? ,F: December twenty-third, please. ,M: How long do you plan to stay? ,F: Four days. ,M: Could you give me your name and phone number? ,F: Nancy Brown, my phone number is 7096785634, Nancy Brown. ,M: Thanks, what time will you arrive? ,F: Around 3:00 PM. ,M: Very good, your reservation confirmation number is NB 13579. Thank you for booking in our hotel. Can I help you with anything else today? ,F: No, that's it. Thank you very much.
(A) M: Okay. Your reservation confirmation number is NB 23579. Now you can go to your room in this hotel. (B) M: You are welcome. As you plan to stay five days, you can enjoy all the food in this restaurant. (C) M: I didnât hear you. Please could you tell me again? (D) M: Okay. You mean you will arrive around 4:00 PM? If you have any question about the room, you can tell me on this phone.
Solution:
|
C
|
task611_mutual_multi_turn_dialogue
|
NIv2
|
fs_opt
| 6
|
train
|
Detailed Instructions: 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.
See one example below:
Problem: M: Excuse me. How much is this suit? ,F: It's on sale today for $750. It's normally $900. ,M: Wow, that is pretty expensive! I was thinking that it might be 4 or 500. ,F: This material is imported from Italy. It's the finest in the world, and if you bought a suit made of this material at many department stores, you would pay about $2000. ,M: Uh-hah. But isn't that the point of coming to a market like this, to get a discount compared to the expensive department stores? Besides I saw a suit just like this one a few stalls down, and they were selling it for $600. I still thought that it was too expensive. ,F: It's possible that the suit you saw was the same color is this one but made of a different material. Unfortunately, our prices are final, and we think there are the lowest anywhere in the city.
(A) M: Okay, I'll buy it at the price you want. Here, it's $600. (B) M: Just a minute! I do not quite follow what you are saying, would you mind repeating that? (C) M: Although the material is imported from France, it is not worth the $750 you want. (D) M: You want $900 for the suit? It's too expensive!
Solution: B
Explanation: Options (A) and (B) are not corrct because the suit's price is 900. Option (C) isn't correct because the material was imported from italy.
Problem: F: You looked tired today, John. What's wrong? ,M: I have so many things to do. I am so stressed out. ,F: Well, just relax. You told me that you have one day off today. You said you would take our daughter to the park. ,M: Yeah, but all I can think about is the deadlines I need to meet. ,F: What deadlines? ,M: Well, I have to have the meeting room organized by Friday. ,F: What else? ,M: I also need to come up with a marketing plan for our dairy products and sign a contract with the sales department by Saturday. ,F: It sounds to me that your boss is just taking too much advantage of you.
(A) M: Youstill think he is helpful after he asked me to organize the meeting room by Thursday. (B) M: I didnât hear you. Please could you tell me again? (C) M: I think so! But what Can I do? He is theboss after all! And a cruel boss! (D) M: No, he is not hard-working at all. He gave all his work to me. But I am still feel relax.
Solution:
|
C
|
task611_mutual_multi_turn_dialogue
|
NIv2
|
fs_opt
| 4
|
train
|
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.
[Q]: F: Oh, it's so nice to be out of the cinema. It was so hot in there. I thought I melt. ,M: Terrible, wasn't it? Well, what did you think of Madame Curie? Do you think it is as good as the review said? ,F: It's a good film, but not as good as the book. I think some of the story was missing in the film. ,M: Yes, but you couldn't possibly fit everything that happens in a 200 page novel into a 2 hour film. There simply isn't the time. ,F: I know, but the director should at least try to follow the story. With this film, he left out some important people and even changed the end. ,M: Well, perhaps he thought he was improving the story, Karen.
(A) F: He did. His film is far more realistic than the book. I like it. (B) F: I am really sorry, I did not catch that. Would you slow down, please? (C) F: Obviouesly, the book is better than the film, though he changed nothing. (D) F: He might. But I still like the original book than the film.
[A]: D
[Q]: F: Is this Mister Brown's office? ,M: Yes, but he's gone out. Did you tell him beforehand about your coming? ,F: Yes, I found him yesterday and he told me to come here at 9:00 today, it's almost the time now. How soon will he be back? ,M: Well, maybe before 10:30. In fact, I'm waiting to see him too. When I arrived at about 8:00, only his secretary was here. She's gone to the copy shop to have a form copied.
(A) F: Iâm sorry, I didnât understand. Could you repeat a little louder, please? (B) F: What? You mean you saw his secretary went out to apply to study abroad at about 9:00? (C) F: I wonder why she went out to copy at 9:00. Don't they have a copying machine in their office? (D) F: But why did she have to go out to contact Mr. Brown? He is in his office for the whole morning.
[A]: A
[Q]: M: My boss told me never to go to work again! ,F: I can't believe it, why? ,M: I don't know. What he told me that I was so confused.
(A) F: Yeah. I can tell that you understood it very well. You will be given a prize soon. (B) F: Excuse me, could you repeat the question? (C) F: Yeah. I can tell that you thought it completely wrong. Actually, you will be given a prize soon. (D) F: Yeah. I can tell that you didn't understand it. Actually, you will be given a prize soon.
[A]:
|
B
|
task611_mutual_multi_turn_dialogue
|
NIv2
|
fs_opt
| 5
|
train
|
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.
Input: Consider Input: F: What do you usually do on Sundays? ,M: I go shopping in the morning, work in the garden in the afternoon and read the newspaper after dinner.
(A) F: Oh, you usually go shopping on Sunday afternoons. (B) F: Oh, you usually work in the garden on Sunday afternoons. (C) F: I am afraid I did not quite catch what you were saying. Please repeat it. (D) F: Oh, you usually read the newspaper on Sunday afternoons.
Output: B
Input: Consider Input: F: Charlie, do you know a restaurant called Bravo? ,M: Bravo? I know the name, but I'm not sure where it is. ,F: It's on George Street. The food there is excellent.
(A) M: Oh, I'm sure where it is. Let's talk more about its dish. (B) M: Excuse me, could you repeat the question? (C) M: Oh, really? It is not on George Street? Let's talk more about this street. (D) M: Yeah, the food there is excellent. Let's talk more about this restaruant.
Output: D
Input: Consider Input: M: Is the meeting held in room 302 or 303? ,F: It should be 302, but I hear that it has been put off till tomorrow. ,M: Really? Is it at 3:00 in the afternoon. ,F: No, it will be at 2:00.
(A) M: OK. The meeting will be held in room 303 at 3:00 pm. I will arrive on time. (B) M: OK. The meeting will be held in room 303 at 2:00 pm. I will arrive on time. (C) M: OK. The meeting will be held in room 302 at 2:00 pm. I will arrive on time. (D) M: Iâm sorry, I didnât understand. Could you repeat a little louder, please?
|
Output: C
|
task611_mutual_multi_turn_dialogue
|
NIv2
|
fs_opt
| 2
|
train
|
Definition: 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.
Input: F: Well, I see you've got your books, George. But are you reading? Looks like you're just coloring. ,M: I'm marking the important parts of the book. The colors are for different things.
(A) F: I didnât hear you. Please could you tell me again? (B) F: Oh, I already knew that you read the book. (C) F: Why do you color in this photo, George? (D) F: Why do you write so many notes in the book?
Output:
|
A
|
task611_mutual_multi_turn_dialogue
|
NIv2
|
zs_opt
| 2
|
test
|
Detailed Instructions: 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.
See one example below:
Problem: M: Excuse me. How much is this suit? ,F: It's on sale today for $750. It's normally $900. ,M: Wow, that is pretty expensive! I was thinking that it might be 4 or 500. ,F: This material is imported from Italy. It's the finest in the world, and if you bought a suit made of this material at many department stores, you would pay about $2000. ,M: Uh-hah. But isn't that the point of coming to a market like this, to get a discount compared to the expensive department stores? Besides I saw a suit just like this one a few stalls down, and they were selling it for $600. I still thought that it was too expensive. ,F: It's possible that the suit you saw was the same color is this one but made of a different material. Unfortunately, our prices are final, and we think there are the lowest anywhere in the city.
(A) M: Okay, I'll buy it at the price you want. Here, it's $600. (B) M: Just a minute! I do not quite follow what you are saying, would you mind repeating that? (C) M: Although the material is imported from France, it is not worth the $750 you want. (D) M: You want $900 for the suit? It's too expensive!
Solution: B
Explanation: Options (A) and (B) are not corrct because the suit's price is 900. Option (C) isn't correct because the material was imported from italy.
Problem: M: Hello. Tenants Advocacy Resource Center. ,F: Hello. I'm having a problem with my house owner. The House owner is a nice enough guy, but he and I just can't seem to agree on repair costs.
(A) M: I' m sorry. Are you just calling to complain about you husband? (B) M: Welcome to the government hall. What kind of help do you want? (C) M: Please clarify your problem further, we will try our best to help you. (D) M: Just a minute! I do not quite follow what you are saying, would you mind repeating that?
Solution:
|
C
|
task611_mutual_multi_turn_dialogue
|
NIv2
|
fs_opt
| 4
|
validation
|
You will be given a definition of a task first, then some input of the task.
Given a premise, an initial context, an original ending, and a counterfactual context, the task is to generate a new story ending aligned with the counterfactual context and as close to the original ending as possible. Each instance consists of a five-sentence story. The premise is the first sentence of a story, and the second sentence, which is the initial context, provides more information about the story's context and the story's general plot. The original ending is the last three sentences of the story. Also, a counterfactual context is a slight modification to the initial context. You should write a new story ending that edits the original story ending as little as possible to regain coherence with the counterfactual context. To sum up, you should write the last three sentences of a story based on the premise(first sentence) and the counterfactual context(second sentence) of the story.
Premise: Kate could not cook.
Initial Context: However, she was determined to cook her boyfriend dinner.
Original Ending: Her mom gave her a great recipe. Kate bought all the ingredients and followed them exactly. The dinner she cooked was perfect!
Counterfactual Context: So she preferred get food from restaurants daily.
Output:
|
Her mom gave her a great recipe the one time she wanted to cook her own dinner. Kate bought all the ingredients and followed them exactly. The dinner she cooked was perfect!
|
task269_csrg_counterfactual_story_generation
|
NIv2
|
zs_opt
| 1
|
train
|
Given a premise, an initial context, an original ending, and a counterfactual context, the task is to generate a new story ending aligned with the counterfactual context and as close to the original ending as possible. Each instance consists of a five-sentence story. The premise is the first sentence of a story, and the second sentence, which is the initial context, provides more information about the story's context and the story's general plot. The original ending is the last three sentences of the story. Also, a counterfactual context is a slight modification to the initial context. You should write a new story ending that edits the original story ending as little as possible to regain coherence with the counterfactual context. To sum up, you should write the last three sentences of a story based on the premise(first sentence) and the counterfactual context(second sentence) of the story.
Premise: I was drinking soda with my friends.
Initial Context: Suddenly, I felt something in my stomach.
Original Ending: I was scared that I would fart. Luckily, it was a burp. Alas, my friends still made fun of me for burping.
Counterfactual Context: Suddenly i had the urge to throw up.
|
I was scared that I would throw up on my friends. Luckily, it was a burp. Alas, my friends still made fun of me for burping.
|
task269_csrg_counterfactual_story_generation
|
NIv2
|
zs_opt
| 0
|
train
|
Given a premise, an initial context, an original ending, and a counterfactual context, the task is to generate a new story ending aligned with the counterfactual context and as close to the original ending as possible. Each instance consists of a five-sentence story. The premise is the first sentence of a story, and the second sentence, which is the initial context, provides more information about the story's context and the story's general plot. The original ending is the last three sentences of the story. Also, a counterfactual context is a slight modification to the initial context. You should write a new story ending that edits the original story ending as little as possible to regain coherence with the counterfactual context. To sum up, you should write the last three sentences of a story based on the premise(first sentence) and the counterfactual context(second sentence) of the story.
Input: Consider Input: Premise: Rebecca had no friends.
Initial Context: She worked and worked to invent a robot best friend.
Original Ending: After Years of laboring, she was successful! Rebecca named the robot Janelle. They were best friends.
Counterfactual Context: She worked hard on her social skills.
Output: After years of improving her social skills, she made friends! Rebecca had a friend named Janelle. They were best friends.
Input: Consider Input: Premise: Two spies met in a wooded clearing.
Initial Context: They came at each other with guns drawn.
Original Ending: Each spy threw down an envelope in front of the other. They lowered their guns and picked up the goods. The spies nodded at each other and departed.
Counterfactual Context: They came at each other shooting wildly.
Output: Each spy threw down their empty clip in front of the other. They lowered their guns and grabbed a new clip. The spies nodded at each other and continued shooting.
Input: Consider Input: Premise: Josh broke up with Hayley years ago.
Initial Context: He wonders if it's a mistake.
Original Ending: He tries to call her and get back with her. All he gets is the dial tone. He figures she's happy with someone else.
Counterfactual Context: He thanks God every day that he's no longer with Hayley.
|
Output: She tries to call and get back together with him. He ignores the phone calls. He's happier with someone else.
|
task269_csrg_counterfactual_story_generation
|
NIv2
|
fs_opt
| 2
|
train
|
Given a premise, an initial context, an original ending, and a counterfactual context, the task is to generate a new story ending aligned with the counterfactual context and as close to the original ending as possible. Each instance consists of a five-sentence story. The premise is the first sentence of a story, and the second sentence, which is the initial context, provides more information about the story's context and the story's general plot. The original ending is the last three sentences of the story. Also, a counterfactual context is a slight modification to the initial context. You should write a new story ending that edits the original story ending as little as possible to regain coherence with the counterfactual context. To sum up, you should write the last three sentences of a story based on the premise(first sentence) and the counterfactual context(second sentence) of the story.
Example: Premise: Susie was sitting on her barstool.
Initial Context: She kept kicking the counter with her feet.
Original Ending: Suddenly, her kick sent her falling backwards. The chair hit the ground with a thud and broke. Susie hurt her head and was really scared.
Counterfactual Context: She kept herself steady with her feet.
Example solution: Suddenly, an earthquake sent her falling backwards. The chair hit the ground with a thud and broke. Susie hurt her head and was really scared.
Example explanation: The generated new ending is perfect. It considers the counterfactual context and changes required parts in original ending.
Problem: Premise: The parents called up the stairs to their daughter.
Initial Context: She immediately came flying down the stairs.
Original Ending: She stood in awe, looking at all the presents. She tore into the gifts and opened them. She promised herself to open them more slowly next year.
Counterfactual Context: She immediately ignored them.
|
Solution: The parents got frustrated because they had bought her presents. They decided to give her one more chance. She didn't come and so they decided not to give her the presents.
|
task269_csrg_counterfactual_story_generation
|
NIv2
|
fs_opt
| 5
|
train
|
Given a premise, an initial context, an original ending, and a counterfactual context, the task is to generate a new story ending aligned with the counterfactual context and as close to the original ending as possible. Each instance consists of a five-sentence story. The premise is the first sentence of a story, and the second sentence, which is the initial context, provides more information about the story's context and the story's general plot. The original ending is the last three sentences of the story. Also, a counterfactual context is a slight modification to the initial context. You should write a new story ending that edits the original story ending as little as possible to regain coherence with the counterfactual context. To sum up, you should write the last three sentences of a story based on the premise(first sentence) and the counterfactual context(second sentence) of the story.
--------
Question: Premise: Sam liked Sally very much, but he was nervous.
Initial Context: He'd never been on a date with a girl before.
Original Ending: He practiced asking her out over and over. Finally he approached her and asked for his date. To his joy, she eagerly agreed!
Counterfactual Context: Sally was always really rude to him and Sam didn't know if she would like him.
Answer: He practiced telling her off over and over. Finally he approached her and asked for her to stop. To his joy, she finally agreed!
Question: Premise: Tina had to take her daughters to school.
Initial Context: She had to leave early because work was early.
Original Ending: She dropped them off an hour early. When the kids got there no one was there. They felt lonely.
Counterfactual Context: She left a little late because her work was starting lat.
Answer: She dropped them off an hour late. When the kids got there everyone was in class. They felt embarrassed.
Question: Premise: Nancy wanted to see a movie, but she had no one to go with.
Initial Context: She had asked all of her friends, so she decided to just go alone.
Original Ending: At first she was uncomfortable, sitting amongst the crowd on her own. As the movie went on, she soon forgot everything that bothered her. Nancy was happy to learn, going alone wasn't so bad after all!
Counterfactual Context: She decided to just stream one instead.
Answer:
|
At first she couldn't find anything but then she found one. As the movie went on she started enjoying the movie. Nancy was happy that she had decided to stay home.
|
task269_csrg_counterfactual_story_generation
|
NIv2
|
fs_opt
| 7
|
train
|
Given a premise, an initial context, an original ending, and a counterfactual context, the task is to generate a new story ending aligned with the counterfactual context and as close to the original ending as possible. Each instance consists of a five-sentence story. The premise is the first sentence of a story, and the second sentence, which is the initial context, provides more information about the story's context and the story's general plot. The original ending is the last three sentences of the story. Also, a counterfactual context is a slight modification to the initial context. You should write a new story ending that edits the original story ending as little as possible to regain coherence with the counterfactual context. To sum up, you should write the last three sentences of a story based on the premise(first sentence) and the counterfactual context(second sentence) of the story.
Q: Premise: I smelled something foul in the house.
Initial Context: I decided to investigate.
Original Ending: I followed the smell up to the attic. I found out that the smell was a dead animal. I disposed of the animal and my house no longer smelled foul.
Counterfactual Context: I found a rotten potato under the sink.
A: I actually vomited because of the smell. I put the potato on a garbage bag. I quickly disposed the rotten potato and my house is no longer smelling foul.
****
Q: Premise: All of the litter of puppies had gone to an new home except one.
Initial Context: Because it was the runt of the litter nobody wanted her.
Original Ending: The people were trying to give her away outside of the grocery store. Since nobody wanted her a kind woman decided to take the puppy home. Finally the unwanted puppy had a home.
Counterfactual Context: Because it was the runt of the litter and had died.
A: The people were trying to bury her outside of the grocery store. Since nobody wanted to help a kind woman decided to help. Finally the runt had a grave.
****
Q: Premise: Christine is heading to warner bros records.
Initial Context: She is meeting up with the labels boss.
Original Ending: The boss asks her to sign the contract. She reads it and signs. She is happy that she is finally signed.
Counterfactual Context: She is picking up her boyfriend from a meeting with the labels boss.
A:
|
The boss asked her boyfriend to sign a contract. He reads it and sign it. She is happy that he is finally signed.
****
|
task269_csrg_counterfactual_story_generation
|
NIv2
|
fs_opt
| 4
|
train
|
Definition: Given a premise, an initial context, an original ending, and a counterfactual context, the task is to generate a new story ending aligned with the counterfactual context and as close to the original ending as possible. Each instance consists of a five-sentence story. The premise is the first sentence of a story, and the second sentence, which is the initial context, provides more information about the story's context and the story's general plot. The original ending is the last three sentences of the story. Also, a counterfactual context is a slight modification to the initial context. You should write a new story ending that edits the original story ending as little as possible to regain coherence with the counterfactual context. To sum up, you should write the last three sentences of a story based on the premise(first sentence) and the counterfactual context(second sentence) of the story.
Input: Premise: Steele was a really good athlete.
Initial Context: He had a big football game coming up.
Original Ending: Steele practiced for his game tirelessly for a month. On the day of the big game, Steele was very prepared. Steele kicked the winning goal for his team.
Counterfactual Context: He wouldn't be able to play in the football game coming up because of a broken arm.
Output:
|
Steele watched as his team practiced , while he sat tirelessly for a month. On the day of the big game, Steele was cheering the team on. Steele watched as his backup kicked the winning goal for his team
|
task269_csrg_counterfactual_story_generation
|
NIv2
|
zs_opt
| 2
|
train
|
TASK DEFINITION: Given a premise, an initial context, an original ending, and a counterfactual context, the task is to generate a new story ending aligned with the counterfactual context and as close to the original ending as possible. Each instance consists of a five-sentence story. The premise is the first sentence of a story, and the second sentence, which is the initial context, provides more information about the story's context and the story's general plot. The original ending is the last three sentences of the story. Also, a counterfactual context is a slight modification to the initial context. You should write a new story ending that edits the original story ending as little as possible to regain coherence with the counterfactual context. To sum up, you should write the last three sentences of a story based on the premise(first sentence) and the counterfactual context(second sentence) of the story.
PROBLEM: Premise: I read about a really great investment and decided to take advantage.
Initial Context: I invested $10,000 with the company.
Original Ending: A couple of months later, I read that it was a Ponzi scheme. I lost all the money I had invested in it. The person running the company went to prison.
Counterfactual Context: I invested $0 with the company.
SOLUTION: A couple of months later, I read that it was a Ponzi scheme. I was glad I hadn't invest money in it. The person running the company went to prison.
PROBLEM: Premise: Sandy and her family live right next to the river.
Initial Context: During a powerful storm, the river rose extremely high.
Original Ending: Sandy kept an eye on the water as it inched closer to her home. She knew they needed to evacuate before they had no way out. Taking very things, they quickly left as water seeped into their home.
Counterfactual Context: During a rainy storm, the river luckily did not rise very high.
SOLUTION: Sandy kept an eye on the water in case it inched closer to her home. She knew if it did they would need to evacuate before they had no way out. Packing all their things, they watched in case water seeped into their home.
PROBLEM: Premise: Dan was recently taking nature pictures in his backyard.
Initial Context: While taking a picture of a tree, his daughter ran across the yard.
Original Ending: Dan wanted just the tree in the picture so he tried again. While looking through the pictures, Dan saw something funny. Dan noticed his daughter had photo bombed the second tree photo.
Counterfactual Context: He got a perfect unobstructed photo of the tree.
SOLUTION:
|
Dan wanted to get his daughter in the picture, so he tried again. While looking through the pictures, Dan saw something funny. Dan noticed his dog had photo bombed the second tree photo.
|
task269_csrg_counterfactual_story_generation
|
NIv2
|
fs_opt
| 8
|
train
|
Teacher: Given a premise, an initial context, an original ending, and a counterfactual context, the task is to generate a new story ending aligned with the counterfactual context and as close to the original ending as possible. Each instance consists of a five-sentence story. The premise is the first sentence of a story, and the second sentence, which is the initial context, provides more information about the story's context and the story's general plot. The original ending is the last three sentences of the story. Also, a counterfactual context is a slight modification to the initial context. You should write a new story ending that edits the original story ending as little as possible to regain coherence with the counterfactual context. To sum up, you should write the last three sentences of a story based on the premise(first sentence) and the counterfactual context(second sentence) of the story.
Teacher: Now, understand the problem? If you are still confused, see the following example:
Premise: Susie was sitting on her barstool.
Initial Context: She kept kicking the counter with her feet.
Original Ending: Suddenly, her kick sent her falling backwards. The chair hit the ground with a thud and broke. Susie hurt her head and was really scared.
Counterfactual Context: She kept herself steady with her feet.
Solution: Suddenly, an earthquake sent her falling backwards. The chair hit the ground with a thud and broke. Susie hurt her head and was really scared.
Reason: The generated new ending is perfect. It considers the counterfactual context and changes required parts in original ending.
Now, solve this instance: Premise: Isabelle was very proud of her long hair.
Initial Context: She took good care of them until she slept on the grass on her lawn.
Original Ending: When she woke up, she realized her hair was full of chewing gum. She cried and went to her mom. Her mom cut all of her hair until it was short to the ears.
Counterfactual Context: She took terrible of her hair and had to chop it short.
Student:
|
When she woke up, she realized how much she missed her long hair. She cried and went to her mom. Her mom helped her take better care of her hair until it grew out again.
|
task269_csrg_counterfactual_story_generation
|
NIv2
|
fs_opt
| 2
|
test
|
Given a premise, an initial context, an original ending, and a counterfactual context, the task is to generate a new story ending aligned with the counterfactual context and as close to the original ending as possible. Each instance consists of a five-sentence story. The premise is the first sentence of a story, and the second sentence, which is the initial context, provides more information about the story's context and the story's general plot. The original ending is the last three sentences of the story. Also, a counterfactual context is a slight modification to the initial context. You should write a new story ending that edits the original story ending as little as possible to regain coherence with the counterfactual context. To sum up, you should write the last three sentences of a story based on the premise(first sentence) and the counterfactual context(second sentence) of the story.
Premise: Last month I took my very first trip to New York City.
Initial Context: I was super excited to stay near Central Park.
Original Ending: The trip was going well until I got bitten by a stray dog. I had to be rushed to the hospital. Thankfully I was alright but it was still scary.
Counterfactual Context: Unfortunately, a snowstorm cancelled my flight.
|
The trip was still going to happen as I caught the next flight. I had to be rushed to the next gate. Thankfully I was on time but it was still scary.
|
task269_csrg_counterfactual_story_generation
|
NIv2
|
zs_opt
| 0
|
validation
|
Detailed Instructions: In this task, you're given a text and question. The question is about the topic of the document. You are expected to generate an incorrect answer. The incorrect answer should be among one of the following categories: 1)Company, 2)Educational Institution, 3)Artist, 4)Athlete, 5)Office holder, 6)Mean of transportation, 7)Building, 8)Natural place, 9)Village, 10)Animal, 11)Plant, 12)Album, 13)Film, 14)Written work
Problem:Text: Western Sierra Law School (WSLS) is a private independent and non-sectarian postsecondary degree granting institution located in San Diego California.
Question: WHat is the classified topic name
Solution:
|
Village
|
task631_dbpedia_14_incorrect_answer_generation
|
NIv2
|
zs_opt
| 8
|
train
|
Definition: In this task, you're given a text and question. The question is about the topic of the document. You are expected to generate an incorrect answer. The incorrect answer should be among one of the following categories: 1)Company, 2)Educational Institution, 3)Artist, 4)Athlete, 5)Office holder, 6)Mean of transportation, 7)Building, 8)Natural place, 9)Village, 10)Animal, 11)Plant, 12)Album, 13)Film, 14)Written work
Input: Text: Tillandsia subinflata is a species of the genus Tillandsia. This species is endemic to Mexico.
Question: WHat is the classified topic name
Output:
|
Company
|
task631_dbpedia_14_incorrect_answer_generation
|
NIv2
|
zs_opt
| 2
|
train
|
Given the task definition and input, reply with output. In this task, you're given a text and question. The question is about the topic of the document. You are expected to generate an incorrect answer. The incorrect answer should be among one of the following categories: 1)Company, 2)Educational Institution, 3)Artist, 4)Athlete, 5)Office holder, 6)Mean of transportation, 7)Building, 8)Natural place, 9)Village, 10)Animal, 11)Plant, 12)Album, 13)Film, 14)Written work
Text: Maureen Elizabeth Murphy (born March 23 1939) is an American former competition swimmer who represented the United States at the 1956 Summer Olympics in Melbourne Australia. Murphy competed in the women's 100-meter backstroke and finished fifth overall in the event final with a time of 1:14.1.
Question: WHat is the classified topic name
|
Artist
|
task631_dbpedia_14_incorrect_answer_generation
|
NIv2
|
zs_opt
| 5
|
train
|
Detailed Instructions: In this task, you're given a text and question. The question is about the topic of the document. You are expected to generate an incorrect answer. The incorrect answer should be among one of the following categories: 1)Company, 2)Educational Institution, 3)Artist, 4)Athlete, 5)Office holder, 6)Mean of transportation, 7)Building, 8)Natural place, 9)Village, 10)Animal, 11)Plant, 12)Album, 13)Film, 14)Written work
See one example below:
Problem: Text: The Mansfield Building Society is a UK building society which has its headquarters in Mansfield Nottinghamshire. It is a member of the Building Societies Association.
Question: What is the classified topic of the document?
Solution: Company
Explanation: "Mansfield Building Society" is a UK building society. Hence, the output is incorrect answer.
Problem: Text: Paal Frisvold (born 5 May 1962) is a Norwegian organizational leader and retired fencer.He competed in the team épée event at the 1984 Summer Olympics. In 2009 he was elected as the new leader of the European Movement in Norway succeeding Svein Roald Hansen.
Question: WHat is the classified topic name
Solution:
|
Building
|
task631_dbpedia_14_incorrect_answer_generation
|
NIv2
|
fs_opt
| 4
|
train
|
In this task, you're given a text and question. The question is about the topic of the document. You are expected to generate an incorrect answer. The incorrect answer should be among one of the following categories: 1)Company, 2)Educational Institution, 3)Artist, 4)Athlete, 5)Office holder, 6)Mean of transportation, 7)Building, 8)Natural place, 9)Village, 10)Animal, 11)Plant, 12)Album, 13)Film, 14)Written work
Q: Text: Pteroteinon capronnieri Capronnierâs Red-eye is a butterfly in the Hesperiidae family. It is found in Ivory Coast Ghana Nigeria Cameroon the Republic of Congo the Central African Republic the Democratic Republic of Congo western Uganda and north-western Tanzania. The habitat consists of deep forests.
Question: WHat is the classified topic name
A:
|
Natural place
|
task631_dbpedia_14_incorrect_answer_generation
|
NIv2
|
zs_opt
| 4
|
train
|
In this task, you're given a text and question. The question is about the topic of the document. You are expected to generate an incorrect answer. The incorrect answer should be among one of the following categories: 1)Company, 2)Educational Institution, 3)Artist, 4)Athlete, 5)Office holder, 6)Mean of transportation, 7)Building, 8)Natural place, 9)Village, 10)Animal, 11)Plant, 12)Album, 13)Film, 14)Written work
One example is below.
Q: Text: The Mansfield Building Society is a UK building society which has its headquarters in Mansfield Nottinghamshire. It is a member of the Building Societies Association.
Question: What is the classified topic of the document?
A: Company
Rationale: "Mansfield Building Society" is a UK building society. Hence, the output is incorrect answer.
Q: Text: Thalastaanam (The Capital) is a 1992 Malayalam film by Shaji Kailas starring Suresh Gopi Narendra Prasad Geetha M. G. Soman and Vijayakumar. Vijayakumar made his debut through this movie.
Question: WHat is the classified topic name
A:
|
Written work
|
task631_dbpedia_14_incorrect_answer_generation
|
NIv2
|
fs_opt
| 9
|
train
|
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're given a text and question. The question is about the topic of the document. You are expected to generate an incorrect answer. The incorrect answer should be among one of the following categories: 1)Company, 2)Educational Institution, 3)Artist, 4)Athlete, 5)Office holder, 6)Mean of transportation, 7)Building, 8)Natural place, 9)Village, 10)Animal, 11)Plant, 12)Album, 13)Film, 14)Written work
Text: The Mansfield Building Society is a UK building society which has its headquarters in Mansfield Nottinghamshire. It is a member of the Building Societies Association.
Question: What is the classified topic of the document?
Solution: Company
Why? "Mansfield Building Society" is a UK building society. Hence, the output is incorrect answer.
New input: Text: The first USS Yorktown was a 16-gun sloop-of-war of the United States Navy. Used mostly for patrolling in the Pacific and anti-slave trade duties in African waters the vessel was wrecked off Maio Cape Verde in 1850.
Question: WHat is the classified topic name
Solution:
|
Building
|
task631_dbpedia_14_incorrect_answer_generation
|
NIv2
|
fs_opt
| 0
|
train
|
Q: In this task, you're given a text and question. The question is about the topic of the document. You are expected to generate an incorrect answer. The incorrect answer should be among one of the following categories: 1)Company, 2)Educational Institution, 3)Artist, 4)Athlete, 5)Office holder, 6)Mean of transportation, 7)Building, 8)Natural place, 9)Village, 10)Animal, 11)Plant, 12)Album, 13)Film, 14)Written work
Text: Eyvind Johnson (29 July 1900 â 25 August 1976) was a Swedish writer and author. He became a member of the Swedish Academy in 1957 and shared the Nobel Prize in Literature with Harry Martinson in 1974 with the citation: for a narrative art far-seeing in lands and ages in the service of freedom.Johnson was born Olof Edvin Verner Jonsson in Svartbjörnsbyn village in ÃverluleÃ¥ parish near the town of Boden in Norrbotten.
Question: WHat is the classified topic name
A:
|
Educational Institution
|
task631_dbpedia_14_incorrect_answer_generation
|
NIv2
|
zs_opt
| 7
|
train
|
Detailed Instructions: In this task, you're given a text and question. The question is about the topic of the document. You are expected to generate an incorrect answer. The incorrect answer should be among one of the following categories: 1)Company, 2)Educational Institution, 3)Artist, 4)Athlete, 5)Office holder, 6)Mean of transportation, 7)Building, 8)Natural place, 9)Village, 10)Animal, 11)Plant, 12)Album, 13)Film, 14)Written work
Q: Text: RFA Maine was a 7432 GRT hospital ship which was built in 1924 as the ocean liner Leonardo da Vinci by SA Ansaldo La Spezia Italy for the Società di Navigazione Transatlantica Italiana. In 1941 she was captured by the British at Kismayu Italian Somaliland. Declared a prize of war she was passed to the Ministry of War Transport (MoWT) and renamed Empire Clyde serving as a hospital ship for the British Army during the Second World War.
Question: WHat is the classified topic name
A:
|
Animal
|
task631_dbpedia_14_incorrect_answer_generation
|
NIv2
|
zs_opt
| 9
|
test
|
In this task, you're given a text and question. The question is about the topic of the document. You are expected to generate an incorrect answer. The incorrect answer should be among one of the following categories: 1)Company, 2)Educational Institution, 3)Artist, 4)Athlete, 5)Office holder, 6)Mean of transportation, 7)Building, 8)Natural place, 9)Village, 10)Animal, 11)Plant, 12)Album, 13)Film, 14)Written work
Example: Text: The Mansfield Building Society is a UK building society which has its headquarters in Mansfield Nottinghamshire. It is a member of the Building Societies Association.
Question: What is the classified topic of the document?
Example solution: Company
Example explanation: "Mansfield Building Society" is a UK building society. Hence, the output is incorrect answer.
Problem: Text: The European ass (Equus hydruntinus) is an extinct equine from the middle and late Pleistocene of Eurasia. It appeared first in the fossil report 300000 years before present and did not disappear until Holocene times. In the late Pleistocene it was widespread throughout much of western Eurasia from the Middle East to Europe especially along the Mediterranean with fossil reports from Sicily Turkey Spain France and Portugal.
Question: WHat is the classified topic name
|
Solution: Athlete
|
task631_dbpedia_14_incorrect_answer_generation
|
NIv2
|
fs_opt
| 5
|
validation
|
Given a sentence, generate a new 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.
Q: You are likely to find a chess attack in a chess game .
A: You are likely to find a match king in a chess restaurant .
****
Q: The effect of going to work is you earn compensation .
A: The effect of going to work is you earn money.
****
Q: You are ok to fly a rock in a group .
A:
|
You are likely to name a lot in a group .
****
|
task413_mickey_en_sentence_perturbation_generation
|
NIv2
|
fs_opt
| 4
|
train
|
Detailed Instructions: Given a sentence, generate a new 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.
See one example below:
Problem: Stimulating the mind helps the brain to grow essential neural cells .
Solution: Stimulating the learning helps the brain to grow new neural cells .
Explanation: This is a good example of a change in the input. The created sentence is semantically similar to the input as both are talking about growing essential neural cells and the changes in the sentence follows the commonsense knowledge.
Problem: You can use a tray of milk to dip a liquid in .
Solution:
|
You can use a glass of regular to dip a cookie in .
|
task413_mickey_en_sentence_perturbation_generation
|
NIv2
|
fs_opt
| 4
|
train
|
Detailed Instructions: Given a sentence, generate a new 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.
Problem:You are likely to find newspapers at a walking companion .
Solution:
|
You are careful to lack newspapers at a train station .
|
task413_mickey_en_sentence_perturbation_generation
|
NIv2
|
zs_opt
| 8
|
train
|
Given a sentence, generate a new 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.
Q: You would promote a drama because you contributed money to see it .
A:
|
You would rate a commission because you paid money to see it .
|
task413_mickey_en_sentence_perturbation_generation
|
NIv2
|
zs_opt
| 4
|
train
|
You will be given a definition of a task first, then some input of the task.
Given a sentence, generate a new 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.
You would attend by credit card because you have no cash .
Output:
|
You would shop by h uff because you have no cash .
|
task413_mickey_en_sentence_perturbation_generation
|
NIv2
|
zs_opt
| 1
|
train
|
Part 1. Definition
Given a sentence, generate a new 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.
Part 2. Example
Stimulating the mind helps the brain to grow essential neural cells .
Answer: Stimulating the learning helps the brain to grow new neural cells .
Explanation: This is a good example of a change in the input. The created sentence is semantically similar to the input as both are talking about growing essential neural cells and the changes in the sentence follows the commonsense knowledge.
Part 3. Exercise
You can use a chess king to play a multiplayer game .
Answer:
|
You can use a chess king to play a chess game.
|
task413_mickey_en_sentence_perturbation_generation
|
NIv2
|
fs_opt
| 7
|
train
|
Given a sentence, generate a new 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.
Example input: Stimulating the mind helps the brain to grow essential neural cells .
Example output: Stimulating the learning helps the brain to grow new neural cells .
Example explanation: This is a good example of a change in the input. The created sentence is semantically similar to the input as both are talking about growing essential neural cells and the changes in the sentence follows the commonsense knowledge.
Q: You are likely to fire a tree in the kitchen .
A:
|
You are likely to find a cat in the pack .
|
task413_mickey_en_sentence_perturbation_generation
|
NIv2
|
fs_opt
| 3
|
train
|
Given a sentence, generate a new 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.
You are likely to find a pillowcase in a bedroom.
You are likely to visit a pillowcase in a bedroom .
If you beg to awake then you should set up your clock ob .
If you want to awake then you should gear up your dinner alarm .
You are likely to find a round trip base in an airport .
|
You are likely to find a round trip translator in an airport .
|
task413_mickey_en_sentence_perturbation_generation
|
NIv2
|
fs_opt
| 0
|
train
|
Given a sentence, generate a new 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: Stimulating the mind helps the brain to grow essential neural cells .
The answer to this example can be: Stimulating the learning helps the brain to grow new neural cells .
Here is why: This is a good example of a change in the input. The created sentence is semantically similar to the input as both are talking about growing essential neural cells and the changes in the sentence follows the commonsense knowledge.
OK. solve this:
You are likely to find an amateur loaf around in a countryside .
Answer:
|
You are likely to find an apple tree around in a countryside.
|
task413_mickey_en_sentence_perturbation_generation
|
NIv2
|
fs_opt
| 8
|
test
|
Given the task definition and input, reply with output. Given a sentence, generate a new 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.
Information can be included as patterns of magnetic charges .
|
Information can be stored as derivatives of magnetic charges .
|
task413_mickey_en_sentence_perturbation_generation
|
NIv2
|
zs_opt
| 5
|
validation
|
Detailed Instructions: A text is given in English. Translate it from the English language to the Telugu language. The translation must not omit or add information to the original sentence.
Q: and the CEO, National Payment Corporation of India (NPCI) will look into the industry cost structure of such transactions which will form the basis to determine the levels of reimbursement.
A:
|
à°®à°à°€à±à°°à°¿à°€à±à°µ à°¶à°Ÿà° à°à°Ÿà°°à±à°¯âà°Šâà°°à±à°¶à°¿ à°®âరియౠచà±à°·âà°šâలౠపà±à°®à±à°à°à± à°à°Ÿà°°à±à°ªà±à°°à±à°·âà°šà± à°à°«à± à°à°à°¡à°¿à°¯à°Ÿ (à°à°šà±âపిఞిà°) à°®à±à°à±à°¯ à°à°Ÿà°°à±à°¯âà°šà°¿à°°à±à°µâà°¹âà°£ à°
à°§à°¿à°à°Ÿà°°à°¿ (à°žà°¿à°à°) à°²âఀౠà°à±à°¡à°¿à°š à°à° à°žà°à°à° à°ªâà°°à°¿à°¶à±à°²à°¿à°žà±à°€à±à°à°Šà°¿.
|
task1047_pib_translation_english_telugu
|
NIv2
|
zs_opt
| 9
|
train
|
Q: A text is given in English. Translate it from the English language to the Telugu language. The translation must not omit or add information to the original sentence.
Therefore, we are not in the competition.
A:
|
à°
à°à°Šà±à°µà°²à±à°², à°®à±à°®à± à°ªà±à°à±à°²à± à°²à±à°®à±.
|
task1047_pib_translation_english_telugu
|
NIv2
|
zs_opt
| 7
|
train
|
Instructions: A text is given in English. Translate it from the English language to the Telugu language. The translation must not omit or add information to the original sentence.
Input: Measures related to MSMEs:
Output:
|
à°à°¿à°šà±à°š మధà±à°¯à°€à°°à°¹à°Ÿ à°žà°à°žà±à°¥à°² à°à±à°žà° à°à±à°ªà°à±à°à± à°à°°à±à°¯à°²à±:
|
task1047_pib_translation_english_telugu
|
NIv2
|
zs_opt
| 3
|
train
|
Given the task definition, example input & output, solve the new input case.
A text is given in English. Translate it from the English language to the Telugu language. The translation must not omit or add information to the original sentence.
Example: Our nation has an experience of 200-225 years in manufacturing metal
Output: మట à°Šà±à°¶à° ఀయటరà±à°²à± 200-225 à°žà°à°µà°€à±à°žà°°à°Ÿà°²à°²à± à°
à°šà±à°à°µà° à°à°à°Šà°¿
Correct translation for given sentence. Input sentence means 'Our nation has an experience of 200-225 years in manufacturing metal' which is the same as the output sentence.
New input case for you: 7090. 68 crore in the financial year 2018-19 (for a period of 14 months from January, 2018 to February, 2019).
Output:
|
72 à°à±à°à±à°²à± - à°
à°Šà±à°µà°¿à°§à°à°à°Ÿ à° à°à°°à±à°§à°¿à° à°žà°à°µà°€à±à°žà°°à° 2018-19 (2018 à°à°šà°µà°°à°¿ à°šà±à°à°¡à°¿ 2019 à°«à°¿à°¬à±à°°à°µà°°à°¿ వరà°à± 14 à°šà±à°²à°²à°à±) à°°à±|| 7090. 68 à°à±à°à±à°² à°à°Ÿà°°à° పడà±à°€à±à°à°Šà°¿.
|
task1047_pib_translation_english_telugu
|
NIv2
|
fs_opt
| 1
|
train
|
You will be given a definition of a task first, then some input of the task.
A text is given in English. Translate it from the English language to the Telugu language. The translation must not omit or add information to the original sentence.
Ministry of Mines VIKAS PARV ON 1ST JUNE IN HYDERABAD TO COMMEMORATE THE COMPLETION OF 3 YEARS OF NDA RULE AT THE CENTRE
Output:
|
à°à±à°à°Šà±à°°à°à°²à± à°®à±à°Šà± à°ªà±à°°à°à±à°€à±à°µà° 3 à°žà°à°µà°€à±à°žà°°à°Ÿà°² పటచ à°ªà±à°°à±à°€à°¿ à°žà°à°Šà°°à±à°à°à°²à± 2017 à°à±à°šà± 1à°š à°¹à±à°Šà°°à°Ÿà°¬à°Ÿà°Šà± లౠâవిà°à°Ÿà°žà± పరà±à°µà±â à°ªà±à°°à°€à±à°¯à±à° à°à°Ÿà°°à±à°¯à°à±à°°à°®à° - à°à±à°à°Šà±à°° à°®à°à°€à±à°°à°¿ à°¶à±à°°à± à°Šà°€à±à°€à°Ÿà°€à±à°°à±à°¯ à°à±à°€à±à°² à°®à±à°Šà±à°à°Ÿ à° à°à°Ÿà°°à±à°¯à°à±à°°à°®à° à°ªà±à°°à°Ÿà°°à°à°à° à°à°Ÿà°šà±à°à°Šà°¿: à°¶à±à°°à± à°žà±à°µà°ªà°šà± à°à±à°®à°Ÿà°°à± à°Šà°€à±à°€à°Ÿ, à°¡à°¿à°ªà±à°¯à±à°à± à°¡à±à°°à±à°à±à°à°°à± à°à°šà°°à°²à± à°à°«à± à°®à±à°šà±à°žà± à°žà±à°«à±à°à±, à°žà±à°€à± à°žà±à°à°à±à°°à°²à± à°à±à°šà±
|
task1047_pib_translation_english_telugu
|
NIv2
|
zs_opt
| 1
|
train
|
Q: A text is given in English. Translate it from the English language to the Telugu language. The translation must not omit or add information to the original sentence.
So, that year, several events were organized taking everyone together.
A:
|
à°
à°à°Šà±à°µà°²à±à°², à° à°à°¡à°Ÿà°Šà°¿, à°
à°à°Šà°°à°¿ à°šà°¿ à°à°²à±à°ªà±à°à±à°šà°¿, à°
à°šà±à° à°à°Ÿà°°à±à°¯à°à±à°°à°®à°Ÿà°² à°šà± à°šà°¿à°°à±à°µà°¹à°¿à°à°à°Ÿà°®à±.
|
task1047_pib_translation_english_telugu
|
NIv2
|
zs_opt
| 7
|
train
|
A text is given in English. Translate it from the English language to the Telugu language. The translation must not omit or add information to the original sentence.
Decline in the Total Fertility Rate (TFR) to 2.
à°žà°à°€à°Ÿà°š à°žà°Ÿà°«à°²à±à°¯ à°šà°¿à°·à±à°ªâà°€à±à°€à°¿ 2011లౠ2.
The Prime Minister, Shri Narendra Modi, today interacted with young innovators and Start-Up entrepreneurs from across the country, through video bridge.
à°ªà±à°°âà°§à°Ÿà°š à°®à°à°€à±à°°à°¿ à°¶à±à°°à± à°šâà°°à±à°à°Šà±à°° à°®à±à°Šà± à°Šà±à°¶âà°µà±à°¯à°Ÿà°ªà±à°€à°à°à°Ÿ à°à°šà±à°šà°à±à°µà°à°à°¿ à°¯à±à°µ à°šà±à°€âà°š à°à°µà°¿à°·à±à°âà°°à±à°€âà°²âà°€à±à°šà±, à°žà±à°à°Ÿà°°à±à°à±â-à°
పౠà°à°à°€à±à°°âà°ªà±à°°à±à°šà±à°¯à±à°°à± à°²âà°€à±à°šà± à° à°°à±à°à± à°µà±à°¡à°¿à°¯à± à°à°Ÿà°šà±à°«âà°°à±à°šà±à°žà± మటధà±à°¯âమఠఊà±à°µà°Ÿà°°à°Ÿ à°žà°à°à°Ÿà°·à°¿à°à°à°Ÿà°°à±.
ICAI and ICPAK will collaborate on benchmarking initiatives and trainee accountant exchange programmes.
|
à°à±à°šà±à°¯à°Ÿà°€à± à°à°Ÿà°°à°€à±à°¶à°Ÿà°šà°¿à°à°¿ à°à°Ÿà°²à°Ÿà°à°Ÿà°²à° à°šà±à°à°à°¿ వటణిà°à±à°¯ à°žà°à°¬à°à°§à°Ÿà°²à± à°à°šà±à°šà°Ÿà°¯à°¿.
|
task1047_pib_translation_english_telugu
|
NIv2
|
fs_opt
| 0
|
train
|
Instructions: A text is given in English. Translate it from the English language to the Telugu language. The translation must not omit or add information to the original sentence.
Input: I salute and I pay my homage to them.
Output:
|
వటరిà°à°¿ à°¶à°¿à°°à°žà°Ÿà°à°¿à°µà°à°Šà°šà° à°à±à°žà±à°€à± à°¶à±à°°à°Šà±à°§à°Ÿà°à°à°²à°¿ à°à°à°¿à°žà±à°€à±à°šà±à°šà°Ÿà°šà±.
|
task1047_pib_translation_english_telugu
|
NIv2
|
zs_opt
| 3
|
train
|
Instructions: A text is given in English. Translate it from the English language to the Telugu language. The translation must not omit or add information to the original sentence.
Input: use of spacecraft and space systems and ground system and application of space technology.
Output:
|
à°
à°à°€âà°°à°¿à°à±à°· à°šà±à°, à°
à°à°€à°°à°¿à°à±à°· à°µà±à°¯à°µà°žà±à°¥à°²à± à°®âరియౠà°à±à°°à±à°à°¡à± à°žà°¿à°žà±à°à°®à± à°² à°¯à±à°à±à° విచియà±à°à°;
|
task1047_pib_translation_english_telugu
|
NIv2
|
zs_opt
| 3
|
test
|
A text is given in English. Translate it from the English language to the Telugu language. The translation must not omit or add information to the original sentence.
One example: Our nation has an experience of 200-225 years in manufacturing metal
Solution is here: మట à°Šà±à°¶à° ఀయటరà±à°²à± 200-225 à°žà°à°µà°€à±à°žà°°à°Ÿà°²à°²à± à°
à°šà±à°à°µà° à°à°à°Šà°¿
Explanation: Correct translation for given sentence. Input sentence means 'Our nation has an experience of 200-225 years in manufacturing metal' which is the same as the output sentence.
Now, solve this: We have to supply potable drinking water to 15 crore houses in rural areas, construct 1.
Solution:
|
15 à°à±à°à±à°² à°à±à°°à°Ÿà°®à±à°£ à°à±à°¹à°Ÿà°²âà°à± à°žà±à°°âà°à±à°·à°¿à°€ à°®à°à°à°¿à°šà±à°°à± à°žâà°°âà°«âà°°à°Ÿ à°à±à°¯à°Ÿà°²à°¿.
|
task1047_pib_translation_english_telugu
|
NIv2
|
fs_opt
| 6
|
validation
|
question: What is the answer: As in the famous song Girl From Ipanema, in which country is Ipanema?
answer: brazil
question: What is the answer: Somerset Maugham's novel 'The Moon And Sixpence' is basedon the life of which painter?
answer: paul gaugin
question: What is the answer: What style of jazz which began and was developed in the mid 1940s was pioneered by jazz musicians Charlie Parker and Dizzy Gillespie?
answer: be bop
question: What is the answer: What type of soup is a consommé or broth garnished with long thin strips of vegetables?
answer:
|
julienne
|
trivia_qa_rc_1_1_0
|
Flan2021
|
fs_noopt
| 9
|
train
|
Problem: What is the name of the only lake in the Lake District?
======
A: bassenthwaite
Q: What does a predator hunt?
A: prey
Problem: The Centers for Disease Control are headquartered near what major city?
Answer: atlanta
Q: How many species of tarantula actually possess bites which are fatal to humans
-----
A: none
Q: Which British monarch made Leamington Spa royal?
What is the answer?
A: queen victoria
Q: An apple seed contains which poisonous substance?
What is the answer?
A:
|
cyanide
|
trivia_qa_rc_1_1_0
|
Flan2021
|
fs_opt
| 4
|
train
|
Problem: What is the capital of the archipelago in the South Atlantic Ocean called The Falkland Islands?
======
A: stanley
Problem: Blue Castello cheese is made in which country?
======
A: denmark
Problem: Which personality of the American West is buried next to her partner Frank Butler?
======
A: annie oakley
Problem: What comic strip, created by Lee Falk, tells the story of a line of costumed crime fighters who operate from the fictional African country of Bangalla?
======
A:
|
phantom
|
trivia_qa_rc_1_1_0
|
Flan2021
|
fs_noopt
| 0
|
train
|
Problem: What part of a human cell has its own DNA - separate from human DNA?
======
A: mitochondria
Problem: What structure, designed by John Paxton for Prince Albert, was destroyed by fire on the 30th November 1936?
======
A: crystal palace
Problem: April 6, 1896 saw the start of the modern Olympics era when what European city hosted the Games of the 1 Olympiad?
======
A: athens greece
Problem: "Which famous poem begins with the line, ""Upon a midnight dreary, while I pondered weak and weary""?"
======
A:
|
raven by edgar allen poe
|
trivia_qa_rc_1_1_0
|
Flan2021
|
fs_noopt
| 0
|
train
|
What is the answer: What contest, held between 24 teams, was played in Dubai between 5 and 7 March 2009, and was won by Wales?
|
rugby sevens world cup
|
trivia_qa_rc_1_1_0
|
Flan2021
|
zs_opt
| 3
|
train
|
Problem: What was the name of the competitor to the Matchbox die cast car, introduced by Mattel in 1968?
======
A: hot wheels
Problem: According to Coleridge, where did Kubla Khan decree his pleasure dome?
======
A: xanadu
Problem: "If you went to Bizet's ""Carmen"", what would you expect to see?"
======
A: opera
Problem: What country would you travel through in going from Honduras to Costa Rica?
======
A:
|
nicaragua
|
trivia_qa_rc_1_1_0
|
Flan2021
|
fs_noopt
| 0
|
train
|
Write the answer: "Billy Blaze, an 8-year-old genius who constructed a spaceship in his backyard from old soup cans and other household objects called ""The Bean-with-Bacon Megarocket"" puts on his brother's football helmet and becomes what character in a video game series?"
|
commander keen
|
trivia_qa_rc_1_1_0
|
Flan2021
|
zs_noopt
| 2
|
train
|
Answer this question.
Which actress is the mother of actresses Joely and the late Natasha Richardson?
|
vanessa redgrave
|
trivia_qa_rc_1_1_0
|
Flan2021
|
zs_opt
| 4
|
train
|
Mei Li:
Which nursery rhyme is based on the events at Mount Kenmel on the border of France and Belgium??
Shuai Zheng:
grand old duke of york
Mei Li:
The 'Du Maurier Classic' is competed for in which sport??
Shuai Zheng:
women s golf
Mei Li:
The Dragon Boat Festival is held in Taiwan during which month of the year??
Shuai Zheng:
|
may
|
trivia_qa_rc_1_1_0
|
Flan2021
|
fs_opt
| 6
|
test
|
Problem: Religous reformers - which 16th century Oxford scholar was the first to translate the Bible into English from the original Greek texts? He was charged with heresy, strangled and burned at the stake in 1535 by the Roman Catholic Church.
======
A: william tyndale
Problem: In which Scottish city is the football team Saint Johnstone based
======
A: perth
Problem: She was built in 1869 to carry what cargo?
======
A: tea
Problem: "Which common piece of medical equipment is named from two Greek words for ""chest"" and ""to look at"" ?"
======
A:
|
stethoscope
|
trivia_qa_rc_1_1_0
|
Flan2021
|
fs_noopt
| 0
|
validation
|
Detailed Instructions: A question that is free of any grammatical or logcal errors, should be labeled 'yes', otherwise it should be indicated as 'no'. A question is grammatically correct if all its entities i.e. nouns, verbs, adjectives, prepositions, pronouns, adverbs are at appropriate position. A question is logically correct if the semantic makes sense.
Q: Sentence: Every Saturday we would go to the park and play Toodles' favorite game.
Question: When did it become his favorite game?
A:
|
Yes.
|
task021_mctaco_grammatical_logical
|
NIv2
|
zs_opt
| 9
|
train
|
Teacher:A question that is free of any grammatical or logcal errors, should be labeled 'yes', otherwise it should be indicated as 'no'. A question is grammatically correct if all its entities i.e. nouns, verbs, adjectives, prepositions, pronouns, adverbs are at appropriate position. A question is logically correct if the semantic makes sense.
Teacher: Now, understand the problem? Solve this instance: Sentence: He tried and tried to please every person who came to the store, but they all picked smaller puppies.
Question: Did he still plead the next day?
Student:
|
No.
|
task021_mctaco_grammatical_logical
|
NIv2
|
zs_opt
| 6
|
train
|
A question that is free of any grammatical or logcal errors, should be labeled 'yes', otherwise it should be indicated as 'no'. A question is grammatically correct if all its entities i.e. nouns, verbs, adjectives, prepositions, pronouns, adverbs are at appropriate position. A question is logically correct if the semantic makes sense.
--------
Question: Sentence: I suppose he was attracted to the commotion up the hill."".
Question: When was the commotion?
Answer: No.
Question: Sentence: They then took a boat to Africa and Asia, where they went on a trip through the mountains.
Question: What time did the boat arrive?
Answer: Yes.
Question: Sentence: In the end , they drive off , listening to the radio .
Question: How long was the drive?
Answer:
|
Yes.
|
task021_mctaco_grammatical_logical
|
NIv2
|
fs_opt
| 7
|
train
|
Definition: A question that is free of any grammatical or logcal errors, should be labeled 'yes', otherwise it should be indicated as 'no'. A question is grammatically correct if all its entities i.e. nouns, verbs, adjectives, prepositions, pronouns, adverbs are at appropriate position. A question is logically correct if the semantic makes sense.
Input: Sentence: Thats because the sound couldnt travel without air particles to transfer the sound energy.
Question: Can sound still travel tomorrow?
Output:
|
Yes.
|
task021_mctaco_grammatical_logical
|
NIv2
|
zs_opt
| 2
|
train
|
Detailed Instructions: A question that is free of any grammatical or logcal errors, should be labeled 'yes', otherwise it should be indicated as 'no'. A question is grammatically correct if all its entities i.e. nouns, verbs, adjectives, prepositions, pronouns, adverbs are at appropriate position. A question is logically correct if the semantic makes sense.
See one example below:
Problem: Sentence: Jack played basketball after school, after which he was very tired.
Question: How long did Jack play basketball?
Solution: Yes.
Explanation: All entities are in appropriate position and the question makes sense, so its grammatically and logically correct.
Problem: Sentence: During a blackout , the house is attacked by aliens , who had previously killed Joe and abducted Iris .
Question: How long does a blackout last?
Solution:
|
Yes.
|
task021_mctaco_grammatical_logical
|
NIv2
|
fs_opt
| 4
|
train
|
A question that is free of any grammatical or logcal errors, should be labeled 'yes', otherwise it should be indicated as 'no'. A question is grammatically correct if all its entities i.e. nouns, verbs, adjectives, prepositions, pronouns, adverbs are at appropriate position. A question is logically correct if the semantic makes sense.
Input: Consider Input: Sentence: Dam Square was landlocked for the first time in its history.
Question: When is it expected to not be landlocked again?
Output: Yes.
Input: Consider Input: Sentence: Martinez works two jobs as a physician's assistant and supports his elderly parents and 8-year-old son.
Question: What does Martinez do after he works?
Output: Yes.
Input: Consider Input: Sentence: Wallace lives in Athens, W. Va., with her husband Rob, 37, and her daughters Rachel, 11, and Claire, 5.
Question: Is Wallace still married?
|
Output: Yes.
|
task021_mctaco_grammatical_logical
|
NIv2
|
fs_opt
| 2
|
train
|
A question that is free of any grammatical or logcal errors, should be labeled 'yes', otherwise it should be indicated as 'no'. A question is grammatically correct if all its entities i.e. nouns, verbs, adjectives, prepositions, pronouns, adverbs are at appropriate position. A question is logically correct if the semantic makes sense.
One example: Sentence: Jack played basketball after school, after which he was very tired.
Question: How long did Jack play basketball?
Solution is here: Yes.
Explanation: All entities are in appropriate position and the question makes sense, so its grammatically and logically correct.
Now, solve this: Sentence: In the end , they drive off , listening to the radio .
Question: Was the vehicle they were driving built before the trip started?
Solution:
|
No.
|
task021_mctaco_grammatical_logical
|
NIv2
|
fs_opt
| 6
|
train
|
instruction:
A question that is free of any grammatical or logcal errors, should be labeled 'yes', otherwise it should be indicated as 'no'. A question is grammatically correct if all its entities i.e. nouns, verbs, adjectives, prepositions, pronouns, adverbs are at appropriate position. A question is logically correct if the semantic makes sense.
question:
Sentence: Third, no middle ground exists in what Qutb conceived as a struggle between God and Satan.
Question: What happens after the struggle between God and Satan?
answer:
Yes.
question:
Sentence: They are very flat areas and they have very rich soils.
Question: When did the soils become rich>?
answer:
No.
question:
Sentence: Although disruption efforts around the world had achieved some successes, the core of Bin Laden's organization remained intact.
Question: What time of day did the organization have a meeting?
answer:
|
Yes.
|
task021_mctaco_grammatical_logical
|
NIv2
|
fs_opt
| 9
|
train
|
Part 1. Definition
A question that is free of any grammatical or logcal errors, should be labeled 'yes', otherwise it should be indicated as 'no'. A question is grammatically correct if all its entities i.e. nouns, verbs, adjectives, prepositions, pronouns, adverbs are at appropriate position. A question is logically correct if the semantic makes sense.
Part 2. Example
Sentence: Jack played basketball after school, after which he was very tired.
Question: How long did Jack play basketball?
Answer: Yes.
Explanation: All entities are in appropriate position and the question makes sense, so its grammatically and logically correct.
Part 3. Exercise
Sentence: This sediment has been eroded off areas of land from upstream.
Question: When does the erosion occur?
Answer:
|
No.
|
task021_mctaco_grammatical_logical
|
NIv2
|
fs_opt
| 7
|
test
|
instruction:
A question that is free of any grammatical or logcal errors, should be labeled 'yes', otherwise it should be indicated as 'no'. A question is grammatically correct if all its entities i.e. nouns, verbs, adjectives, prepositions, pronouns, adverbs are at appropriate position. A question is logically correct if the semantic makes sense.
question:
Sentence: Dam Square was landlocked for the first time in its history.
Question: How long was it not landlocked?
answer:
No.
question:
Sentence: She began to tell the story of Majestic, the wild horse who could not be calmed.
Question: What time did she tell the story?
answer:
Yes.
question:
Sentence: You can see examples of people with gravitational potential energy in 1.5 Figure below.
Question: What will a typical person do after reviewing the example?
answer:
|
Yes.
|
task021_mctaco_grammatical_logical
|
NIv2
|
fs_opt
| 9
|
validation
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.