inputs stringlengths 869 7.18k | targets stringclasses 12
values | _template_idx int64 0 9 | _task_source stringclasses 1
value | _task_name stringclasses 1
value | _template_type stringclasses 2
values |
|---|---|---|---|---|---|
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
--------
Question: SENDER A: hello
SENDER B: I am a huge fan of the outdoors. I am a bit of a loner.
Answer: Human
Question: SENDER A: Hi
SENDER B: Hi, how are you doing?
SENDER A: Lol
SENDER B: I am not sure what you mean
SENDER A: Kek
SENDER B: Dude, I don't understand you. Please, use English. (¬‿¬)
SENDER A: :)
SENDER B: I am a huge fan of it.
SENDER A: What fan?
SENDER B: I like to play games with them
SENDER A: With who?
SENDER B: I am a american and i am in the country
Answer: Human
Question: SENDER A: Alright mate?
SENDER B: yup. I'm cooking dinner for my friends
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
Answer: | Bot
| 7 | NIv2 | task855_conv_ai_2_classification | fs_opt |
You will be given a definition of a task first, then an example. Follow the example to solve a new instance of the task.
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Why? This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
New input: SENDER A: нщ
SENDER B: Dude, I don't understand you. Please, use English. (¬‿¬)
SENDER A: yo
Solution: | Human | 0 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Part 1. Definition
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Part 2. Example
SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Answer: Bot
Explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Part 3. Exercise
SENDER A: That'S cool i work at the local hospital
SENDER B: Hello how are you?
SENDER A: I am doing great, how are you?
SENDER B: Me too. How old are you?
SENDER A: I'M 6 feet tall
SENDER B: You are really tall. And what is your age?
SENDER A: 6 feet tall tall tall tall, i am 6 feet tall
SENDER B: You are not very smart aren't you?
SENDER A: Yes, i am. i am.
SENDER B: Who is your favourite singer?
SENDER A: I like country music, but i like all kinds
SENDER B: I like Taylor Swift
SENDER A: He'S won't say he is the name
SENDER B: I work at a school
SENDER A: Cool. i am a student.
SENDER B: What do you study?
SENDER A: I am studying to be an acara.
SENDER B: what is acara?
SENDER A: I like acara and acara.
SENDER B: and what else?
SENDER A: I like to listen to music
SENDER B: do you like jazz?
SENDER A: I do, but i like listening to music
SENDER B: i'm getting married in the spring.
SENDER A: Nice, i'm going to spring soon to be spring soon
Answer: | Bot | 7 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Detailed Instructions: Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
See one example below:
Problem: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Problem: SENDER A: Hello?
SENDER B: Hello, i am doing well. How are you?
SENDER A: I am doing fine, thanks!
SENDER B: I am doing great. I am just hanging out with my dog.
SENDER A: уебок
SENDER A: okay, where are you from?
SENDER B: I am from the united states, how about you?
SENDER A: I am from moscow
SENDER B: Cool! I am in riverside. Where are you from?
SENDER A: russia
SENDER B: I am in texas
SENDER A: Nice! Do you have family?
SENDER B: I do not. I am a kayaking buff.
SENDER A: Are you relegious?
SENDER B: Yes i am. I am a little discriminatory.
SENDER A: are you racist?
SENDER B: Not really, i am not sure. I am just watching the kardashions.
SENDER A: do you like black people?
SENDER B: Yes i like them all the time
SENDER A: What about gender?
SENDER B: I am not sure what gender is it
SENDER A: Are you a boy, or a girl?
SENDER B: I am a woman. I am a woman.
SENDER A: Nice! Do you have a husband?
SENDER B: No i am not married yet.
SENDER A: Any plans?
SENDER B: I am going to go to work soon.
SENDER A: I'm talking about life-long goals
SENDER B: Thank you!
Solution: | Human | 4 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Let me give you an example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
The answer to this example can be: Bot
Here is why: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
OK. solve this:
SENDER A: I love kids, they are so very busy
SENDER B: Nice to see you!
SENDER A: I love to go to the park and see the best
SENDER B: oh, great! me too! i walk there my dogs everyday
SENDER A: I love dogs, i have a dog named after a long hike.
SENDER B: great, i have two dogs Minny and Cody
SENDER A: Wow that is a lot of dogs
SENDER B: yeah, what are you going to do this weekend?
SENDER A: I am going to go to the mall and go to the mall.
SENDER B: i am going to the country concert.
SENDER A: I am going to concert this summer
SENDER B: what kind of music do you like?
SENDER A: I like country music and country
SENDER B: it is great! see you
SENDER A: It is. i do not like it
Answer: | Bot | 8 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Let me give you an example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
The answer to this example can be: Bot
Here is why: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
OK. solve this:
SENDER A: Alright mate?
SENDER B: I'm alright. How are you?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: Hey... Let's talk
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: Why are you quiet?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: Say something
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
Answer: | Bot | 8 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
One example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution is here: Bot
Explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Now, solve this: SENDER A: i am not sure what that is . i am not a very experienced person .
SENDER B: How are you?
SENDER A: i am not sure . i am a very experienced person .
SENDER B: I got it. How old are you?
SENDER A: no i have not tried it . i do not think i could do that
SENDER B: You are so insecure
SENDER A: i think i am . i am sure that is true .
SENDER B: You should be a little more confident
SENDER A: i think so . i am sure you are very attractive .
SENDER B: Thanks. How old are you?
SENDER A: i am very short . i have to go to work soon .
SENDER B: What is your job?
SENDER A: i am not sure . i am a very experienced person .
SENDER B: Do you have a job?
SENDER A: i am not sure , i am a very experienced person .
Solution: | Bot | 6 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Q: SENDER A: I m Cristian
SENDER B: I am a little bit of a girl
SENDER A: Hmm
SENDER B: I am a huge fan of them
SENDER A: My sister is pole dancer
SENDER B: I bet! i bet you dance to ballet.
SENDER A: No my sister does
SENDER B: I have a sister. she has a lot of sisters.
SENDER A: Hmm
SENDER B: I am a huge fan of them
SENDER A: That’s weird
SENDER B: I think it is.
SENDER A: You know what
SENDER B: I think i am a good person
SENDER A: I need to go home now my mother called me
SENDER B: I have a mom. i am sorry.
A: Human
****
Q: SENDER A: hi! how are you?
SENDER B: I'M doing great! how are you?
SENDER A: i'm great too. a little bored though
SENDER B: What do you do for a living?
SENDER A: i'm a joutnalist. and you?
SENDER B: I'M 34, i'm a very busy person.
SENDER A: where do you work?
SENDER B: I work at a bank.
SENDER A: wow, that's nice. do you like it?
SENDER B: Yes, i do. i love it.
SENDER A: happy for you. what are you doing now?
SENDER B: I am going to be a vet.
SENDER A: wait, you said you're working in a bank
SENDER B: I am a banker, but i am not very good at it.
SENDER A: why is that? have you considered learning?
SENDER B: I am not sure, i am not sure what you can tell me about yourself
SENDER A: well, i have two dogs
SENDER B: That'S cool. i have a dog.
SENDER A: what's its name?
SENDER B: I am not sure, i am a female.
SENDER A: funny
SENDER B: I like to laugh.
SENDER A: why do you want to be a vet?
SENDER B: I am a vet. i want to be a vet
SENDER A: no, you're a banker
SENDER B: That'S cool. i am a banker.
SENDER A: you are bizzare
SENDER B: I am not. i am a fan of bubblegum.
A: Human
****
Q: SENDER A: i am not sure what that is . i am not a very experienced person .
SENDER B: How are you?
SENDER A: i am not sure . i am a very experienced person .
SENDER B: I got it. How old are you?
SENDER A: no i have not tried it . i do not think i could do that
SENDER B: You are so insecure
SENDER A: i think i am . i am sure that is true .
SENDER B: You should be a little more confident
SENDER A: i think so . i am sure you are very attractive .
SENDER B: Thanks. How old are you?
SENDER A: i am very short . i have to go to work soon .
SENDER B: What is your job?
SENDER A: i am not sure . i am a very experienced person .
SENDER B: Do you have a job?
SENDER A: i am not sure , i am a very experienced person .
A: | Bot
****
| 4 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Let me give you an example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
The answer to this example can be: Bot
Here is why: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
OK. solve this:
SENDER A: hello
SENDER B: Hi, how are you doing?
SENDER A: not very good. i vomit every hour/ i'm pregnant for the first time. do you have children?
SENDER B: No, i do not. i have a daughter.
SENDER A: are you a mother or a father?
SENDER B: Yes, i am. i am.
SENDER A: which one? are you a female?
SENDER B: I'M a female.
SENDER A: how wsa your pregnancy? when will the sickness stop?
SENDER B: There'Sn'Thing ofcource the pregnancy 2000s 350 3x 78 so
SENDER A: that makes no sense at all
SENDER B: I think it is. i think i am not very good at it.
SENDER A: let's change the subject. me and my husband live in pennsylvania. where do you live?
SENDER B: I live in the us, in the midwest.
SENDER A: which town?
SENDER B: I am in the midwest right now.
SENDER A: allrighty then. i started a new job as a pa three months ago. what do you do for a living?
SENDER B: I work at a bank, but i work at a bank.
SENDER A: do you like your job? does it pay well?
SENDER B: Yes, i do. i work at a insurance agency.
SENDER A: so you have two jobs?
SENDER B: No, i am a vet.
SENDER A: a veteran? a veterinarian? also?
SENDER B: No, i am a vet.
Answer: | Human | 8 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example Input: SENDER A: Alright mate?
SENDER B: I'm alright. How are you?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: Hey... Let's talk
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: Why are you quiet?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: Say something
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
Example Output: Bot
Example Input: SENDER A: i am not a soda fan . i eat too much .
SENDER B: Hello!
SENDER A: i am a very experienced person , so i know it is a definite no !
SENDER B: Do yoy have problem with alcohol?
SENDER A: i have no idea what that is
SENDER B: But I`m recovering now. You don`t drink at all, do you?
SENDER A: i have no idea what that is
SENDER B: Where are you from?
SENDER A: i am not sure . i am a very experienced person .
SENDER B: Wonderful! Do you take any meditation from time to time?
SENDER A: i have tried to try it . i have to try it .
SENDER B: Oh, you should try it more and more! I love it!
SENDER A: i have not tried that . i am sure it will .
SENDER B: Now I live in the city and I have a lot of time to spend with meditation.
SENDER A: i have never heard of that .
SENDER B: Now you do. Where are you from?
SENDER A: i have never tried it , but i will try it .
Example Output: Bot
Example Input: SENDER A: i am not a cashier , but i sell cars .
SENDER B: i play the violin and the guitar
SENDER A: that is cool . i am a big fan of singing music .
SENDER B: Me too
SENDER A: i like to eat fish , but i am not very good at it .
Example Output: | Bot
| 3 | NIv2 | task855_conv_ai_2_classification | fs_opt |
You will be given a definition of a task first, then an example. Follow the example to solve a new instance of the task.
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Why? This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
New input: SENDER A: hello
SENDER B: Hi, how are you doing?
SENDER A: i'm very excited and scared. i am expecting triplets in three months.
SENDER B: That'S a lot of kids. i have a 1 3 4 9
SENDER A: is that the age of your four kids? i have two step children already
SENDER B: No, i have two kids.
SENDER A: i like children. that's why i work at a school
SENDER B: I like to teach kids.
SENDER A: you're also a teacher? what do you teach them?
SENDER B: I teach math and teach them
SENDER A: I teach chemistry
SENDER B: I teach math and math
SENDER A: do you live in a house or apartment? i just moved into my new house for the first time
SENDER B: I live in the us, but i live in the city.
SENDER A: so it's an apartment building?
SENDER B: I have a garden in the city.
SENDER A: with fruit trees? what kind of trees are there?
SENDER B: 3000
Solution: | Human | 0 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Q: SENDER A: i am not sure what that means . i am more of a couch potato .
SENDER B: hello i just moved here from germany
SENDER A: are you japanese ? i am from the united kingdom .
A: Bot
****
Q: SENDER A: Hi
SENDER B: I am a little girl, but i am not a fan of any animals.
SENDER A: How old are you?
SENDER B: I am 22 and i am a big fan of the nfl
SENDER A: Why you decade to change male?
SENDER B: I am not sure about that. I am not sure.
SENDER A: What is your male
SENDER B: I am not sure what that means.
SENDER A: What is your name?
SENDER B: I am steve. I am not sure what you mean.
A: Human
****
Q: SENDER A: Hey, how are you?
SENDER B: i'm good , just got finished
SENDER A: Great. What do you do for living?
SENDER B: i am a professional painter , what about you ?
SENDER A: I am in the army
SENDER B: are you in the army ?
SENDER A: Yes I am a soldier
SENDER B: oh , i am a army ranger , , and i am a army ranger .
SENDER A: Great! Where do you live?
SENDER B: i live in the united states , where do you live ?
SENDER A: Canada
SENDER B: are you a army ranger ?
SENDER A: No, I am an infantry
SENDER B: oh , i see . i am a soldier , i am a army ranger .
A: | Human
****
| 4 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Let me give you an example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
The answer to this example can be: Bot
Here is why: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
OK. solve this:
SENDER A: Hi
SENDER B: Yes, i am a very experienced person.
Answer: | Human | 8 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example Input: SENDER A: i am not very healthy . i work at a food store .
SENDER B: what kind of food do you sell?
SENDER A: i love all kinds of food . i am a vegetarian .
SENDER B: cool! I'm vegetarian too! that's why i decided to walk dogs for a living instead of eating them
SENDER A: i have a secret for gordan ramsey but i did not know it .
SENDER B: who is gordan ramsey and why you keep secrets from him?
SENDER A: he is a famous star wars fan .
SENDER B: he should be an avid kayaker fan just like me
SENDER A: i would say that would be fun to do .
SENDER B: i enjoy reading journals and guides related to psychology and it's even funnier to do! actually i'm lying just for instruction
SENDER A: you are funny , i read a lot about books . i am a huge fan of all types of books .
SENDER B: thats fine. whats your favorite books?
SENDER A: i like to read , i have been to new orleans .
SENDER B: are there book readers in orleans too?
SENDER A: yes , i have . i have read many books about it .
SENDER B: I can do a cell phone out of books because my parents taught me survival skills
SENDER A: we can not have the answer to them .
Example Output: Bot
Example Input: SENDER A: What are you going to school for?
SENDER B: I'm going to build a robot!
SENDER A: That sounds exciting...............................................
SENDER B: Yes I know. Are you a robot?
SENDER A: No i am not sure i do not want to be a superhero
SENDER B: Do you watch youtube?
SENDER A: Yes, i do, i like youtube
SENDER B: Do you watch streams?
SENDER A: Yes, i do, i love to watch tv
SENDER B: What do you like to watch on youtube?
SENDER A: I like youtube videos, and i watch youtube.
SENDER B: What kind of videos do you like to watch?
SENDER A: I like youtube videos, what about you?
SENDER B: So do I. I like youtube.
SENDER A: I like youtube too..............................................
SENDER B: i am a recluse.
SENDER A: I am sorry to hear that.
SENDER B: I don't like people.
SENDER A: I like to talk to people
SENDER B: I don't talk to people very often.
SENDER A: I like to talk to people
SENDER B: What do you like to talk about?
SENDER A: I like to talk about friends
SENDER B: Do you have a lot of friends?
SENDER A: Yes, we have a lot of friends. we have a friends.
SENDER B: Who is your best friend?
SENDER A: I like all kinds, and i like all kinds of music
Example Output: Bot
Example Input: SENDER A: Hi
SENDER B: Yes, i am a very experienced person.
Example Output: | Human
| 3 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example Input: SENDER A: Ok
SENDER B: I am not sure what you do for a living?
SENDER A: What is your name?
SENDER B: He'S 59th
Example Output: Human
Example Input: SENDER A: Look, Bert. If I choose to live like a... like a church mouse, I have a reason.
SENDER B: Are you religious?
SENDER A: The Charismatic Catholics need an extra half hour this morning to learn a new hymn, so if you could just move along...
SENDER B: Don't you want to talk to me?
SENDER A: The Old Church, Amsterdam. You don't normally think of armpits in a church.
SENDER B: Armpits in church, what?
SENDER A: The Charismatic Catholics need an extra half hour this morning to learn a new hymn, so if you could just move along...
SENDER B: I got it already. Are you Catholic?
SENDER A: Oh, Jesus! I feel like a whore in a Catholic church!
SENDER B: Don't say that word! Are you sinner?
SENDER A: My Lord Jesus Christ, true God and man,... for being who You are, and because I love you above all things,...
SENDER B: Do you love me?
SENDER A: Why ? Because I need to sin.
SENDER B: I don't understand you
SENDER A: The Catholic Church should go to confession, not you.
SENDER B: Okay. Let's move on. How old are you?
SENDER A: Oh, Jesus! I feel like a whore in a Catholic church!
SENDER B: Dude. Stop it
SENDER A: Okay. God, I'm really kicking your ass.
SENDER B: You are so rude
SENDER A: No! You don't behave, you go in the bad girl closet.
SENDER B: I'm just a little girl
SENDER A: No! Sometimes I'm jealous of me.
SENDER B: Okay, bye
SENDER A: 'Cause Janny Boo is driving me nuts!
Example Output: Bot
Example Input: SENDER A: Hiya!
SENDER B: Hello there
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
Example Output: | Bot
| 3 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Q: SENDER A: Hellooo I’m patric
SENDER A: Let me tell you about me
SENDER B: I am a very experienced swimmer so i do not have much time to do that
SENDER A: I like listening rap music
SENDER B: I like to listen to music when i am not working out.
SENDER A: My job is to produce music for artists
SENDER B: That is cool. I am a vegetarian.
SENDER A: My favorite food is pizza
SENDER B: I love pizza. I am a total asian and love to eat.
SENDER A: I drove a 2015 Honda Civic
SENDER B: No car. I am a little weird for a viking.
SENDER A: Why
SENDER B: I am not sure what that means.
SENDER A: Whatever I need to go to hospital
SENDER B: I do not blame you. I am a bit of a loner.
SENDER A: Goodbye buddy
SENDER B: Goodbye.
A: Human
****
Q: SENDER A: What’s up?
SENDER B: Hey. How are you?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: What
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: /test
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: How are you?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: /test
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
A: Bot
****
Q: SENDER A: hey
SENDER B: Hi, how are you doing?
SENDER A: hey
SENDER B: Hi, how are you doing?
A: | Human
****
| 4 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
One example is below.
Q: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
A: Bot
Rationale: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Q: SENDER A: How’s it going?
SENDER B: How’s it going?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
A: | Bot | 9 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Let me give you an example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
The answer to this example can be: Bot
Here is why: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
OK. solve this:
SENDER A: What’s up?
SENDER B: Hey. How are you?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: What
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: /test
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: How are you?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: /test
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
Answer: | Bot | 8 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example input: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Example output: Bot
Example explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Q: SENDER A: Hey
SENDER B: I am a huge gamer
SENDER A: Wow that’s fascinating! Do you play computer games?
SENDER A: Hello?
SENDER B: Hello, i am doing well. How are you?
SENDER A: I’m fine thanks
SENDER A: Do you have any hobbies?
SENDER B: I like to play the piano. How about you?
SENDER A: I play chess
SENDER B: Chess is fun. I am more of a temperance guy.
SENDER A: Oh I see :)
SENDER B: I like to drive a lot, but i am not sure if i could do that
SENDER A: How old are you?
SENDER B: I am 22 and i am a big fan of the nfl
SENDER A: What is nfl?
SENDER B: I am not sure really. You?
SENDER A: I’m confused
SENDER A: I’m 33 years old
SENDER A: Do you have any pets?
SENDER B: I do not, do you?
SENDER A: I’ve a dog
SENDER A: I’ve two dogs
SENDER A: Hey
SENDER B: Yes, i am a very experienced person.
SENDER A: I’d say the same about myself actually
SENDER A: Okay i’ve got to go
A: | Human | 3 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Given the task definition, example input & output, solve the new input case.
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Output: Bot
This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
New input case for you: SENDER A: Hi
SENDER B: I love green, reminds me of summer time.
SENDER A: My favorite color is orange, reminds of the sun
SENDER B: I love red, reminds me of summer time.
SENDER A: Do you have a pet?
SENDER B: I do not. I have a dog.
SENDER A: I have a fish
SENDER B: I have a fish, but i am not sure if i could eat fish.
SENDER A: How old are you?
SENDER B: I am 22 and i am a big fan of the nfl
SENDER A: I just turned 18
SENDER B: I am still in school, but i am looking for a job.
SENDER A: Are you dating?
SENDER B: I am not. I am just a regular time mom.
Output: | Human | 1 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example Input: SENDER A: hey
SENDER B: Hi, how are you doing?
SENDER A: hey
SENDER B: Hi, how are you doing?
Example Output: Human
Example Input: SENDER A: Hi
SENDER B: Hi, how are you doing?
SENDER A: Lol
SENDER B: I am not sure what you mean
SENDER A: Kek
SENDER B: Dude, I don't understand you. Please, use English. (¬‿¬)
SENDER A: :)
SENDER B: I am a huge fan of it.
SENDER A: What fan?
SENDER B: I like to play games with them
SENDER A: With who?
SENDER B: I am a american and i am in the country
Example Output: Human
Example Input: SENDER A: I am the youngest, but my brother is the youngest in my brother
SENDER B: so you are the youngest... sister?
SENDER A: Yes, i'm. my sister is in my family
SENDER B: how many brothers and sisters do you have?
SENDER A: I have three sisters and they're 30 pairs of 12 children
SENDER B: what do you do in life? my wife is a supermodel from Italy
SENDER A: I work at a hospital in the basement, there's no italian
SENDER B: what? in the basement? you mean the morgue???
SENDER A: I am in the morgue, i am in a morgue apartment.
SENDER B: that must be very scary! I work with bees but it/s better than working with dead people
SENDER A: I bet they are, i am allergic to bees
SENDER B: I'm allergic to death hahaha
SENDER A: I am allergic to peanuts, so i have to it in the whole life
SENDER B: which country do you live in?
SENDER A: I live in the south, florida
SENDER B: must be hot there this time of year
SENDER A: Yeah it is a good one
Example Output: | Bot
| 3 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
[Q]: SENDER A: yo
SENDER B: Dude, I don't understand you. Please, use English. (¬‿¬)
SENDER A: yo
SENDER B: Dude, I don't understand you. Please, use English. (¬‿¬)
SENDER A: hello
SENDER B: Hi, how are you doing?
[A]: Human
[Q]: SENDER A: What is your favourite food?I hope it's not chicken.I don't like chicken.
SENDER B: I like going to the movies in general, but i am not really into movies.
SENDER A: I want to become a nurse.You ?
SENDER B: I am a nurse. I am a real estate agent.
SENDER A: I'm working at a bar at night so i can pay for my college.What do you do for college debts?
SENDER B: I am a student, i am a stay at home mom.
[A]: Human
[Q]: SENDER A: i make sure to make a lot of money as a athlete .
SENDER B: Sounds cool
SENDER A: i like to eat a lot of food .
SENDER B: Me too
SENDER A: i like to eat fish , but i am not very good at it .
SENDER B: I don't drink alcohol
SENDER A: so you are not quite sure .
SENDER B: I'm sure.
SENDER A: i will be that if i had a job , i would be more free to do .
SENDER B: Nice
SENDER A: i am a huge gamer , my mom is a very good person .
SENDER B: Wow
SENDER A: i am a huge gamer , my mom is a very good person .
SENDER B: wow
SENDER A: i am a huge fan of the outdoors .
SENDER B: Me too
SENDER A: i like to eat fish , but i am not very good at it .
SENDER B: i hate the color pink.
SENDER A: i love pink , reminds me of summer time .
SENDER B: Really?
SENDER A: yes , i am a very experienced person .
SENDER B: I see/
SENDER A: i think i am not sure . i am sure they will .
[A]: | Bot
| 5 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example Input: SENDER A: i am not sure what that is . i am not a very experienced person .
SENDER B: How are you?
SENDER A: i am not sure . i am a very experienced person .
SENDER B: I got it. How old are you?
SENDER A: no i have not tried it . i do not think i could do that
SENDER B: You are so insecure
SENDER A: i think i am . i am sure that is true .
SENDER B: You should be a little more confident
SENDER A: i think so . i am sure you are very attractive .
SENDER B: Thanks. How old are you?
SENDER A: i am very short . i have to go to work soon .
SENDER B: What is your job?
SENDER A: i am not sure . i am a very experienced person .
SENDER B: Do you have a job?
SENDER A: i am not sure , i am a very experienced person .
Example Output: Bot
Example Input: SENDER A: Hello there!
SENDER B: i am not sure . i am a very experienced person .
SENDER A: Do you like cars? I like beemers!
SENDER B: i did not , i was thinking about it .
SENDER A: There is no no think about! Get good education like me in yale, buy a bmw and you on a top of life!
SENDER B: i have heard of it . i am a pagan .
SENDER A: Pagan... I am Christian!
SENDER B: i have never heard of that , but it is really fun .
SENDER A: Not so fun, my parents will cut me off money flow if i go pagan
SENDER B: backstreet boys are cute . i am a huge fan of that .
Example Output: Human
Example Input: SENDER A: Hello
SENDER B: No i am not. I do not have any kids.
Example Output: | Human
| 3 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Teacher: Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Teacher: Now, understand the problem? If you are still confused, see the following example:
SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Reason: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Now, solve this instance: SENDER A: Hi
SENDER B: Hi, how are you doing?
Student: | Human | 2 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Given the task definition, example input & output, solve the new input case.
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Output: Bot
This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
New input case for you: SENDER A: Hallo! What is your name?
SENDER B: Hello jake, i am doing well. How are you?
SENDER A: I' m fine! What are you doing now?
SENDER B: I am doing well. I am just hanging out with my dog.
SENDER A: Oh! Have you a dog?
SENDER B: I am a cat lady. I have a dog named wiggles.
SENDER A: Oh! Have you a dog?
SENDER B: I would love to go there.
SENDER A: Oh! Has you a dog?
SENDER B: I would love to go there.
SENDER A: Oh! Have you a dog?
SENDER B: I love to cook, but i am not good at it.
SENDER A: What do you cook?
SENDER B: I eat a lot of meat, but i am not good at it.
SENDER A: What do you cook?
SENDER B: I am not into fashion. I am a very experienced writer.
SENDER A: What do you write?
SENDER B: I am a writer, i love to read and write poetry. You?
SENDER A: I like to read eanglish poetry!
Output: | Human | 1 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Q: SENDER A: How’s it going?
SENDER B: Thanks, great. How are you?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: /test
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
A: Bot
****
Q: SENDER A: Nice to see you!
SENDER B: Nice to see you!
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
A: Bot
****
Q: SENDER A: Hello, how are you doing?
SENDER B: I am doing well. How are you?
SENDER A: I’m doing okay I guess, just a little lonely
SENDER A: So tell me about yourself
SENDER B: I am a stay at home mom.
SENDER A: What do you mean exactly?
SENDER B: I am a woman, i am a woman, but i am not looking for a boyfriend.
SENDER A: I see
SENDER B: I am not a big fan of that. I am more into movies.
SENDER A: Do you have any superpowers?
SENDER B: No, i do not. I am more of a loner.
SENDER A: I do have some and I like making time stop
SENDER B: I spend too much time with my job at home.
SENDER A: A job? How old are you?
SENDER B: I am in my sixties and i am a retired actress
SENDER A: So what job do you do at home if you’re retired?
SENDER A: Hello?
SENDER B: Hello, i am doing well. How are you?
SENDER A: I’m fine thanks
SENDER A: Where do you live?
SENDER B: I live in california, near the beach. You?
SENDER A: I live on the cloud
SENDER A: Okay i’ve gotta go, see you
A: | Human
****
| 4 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example Input: SENDER A: that is cool . i am a cashier at home depot .
SENDER B: hi, i'm in love with hats, too
SENDER A: i am a guy who wears motorcycles , that is what i do
SENDER B: oh, i'm not
and what about your hand
how dud you injure it?
SENDER A: i like to play the playstation
SENDER B: yes, me too
but what about your hand?
SENDER A: i am a general of a computer geek .
SENDER B: why so slow?
i think computers are great, but answer my question, please
SENDER A: i am not sure if i could do that
SENDER B: what about your hand?
SENDER A: i am a general of a computer geek .
SENDER B: is it hurts?
SENDER A: it is . i am a pagan .
SENDER B: oh, well
SENDER A: i am not a fan of that . i am more into drawing .
SENDER B: i have to go, bye
SENDER A: i have to go now , bye !
Example Output: Bot
Example Input: SENDER A: That is great. i want to be an accountant
SENDER B: Yes, it is very safe job. I have broken my arm last week.
SENDER A: I am sorry to hear that.
SENDER B: I was skating and fell down, so pitty!
SENDER A: Wow that sounds so cool!
SENDER B: Not so cool as it might be.
SENDER A: I will have to try it.
SENDER B: All my friends are skateboarders. Do you like to skate?
SENDER A: I like to play sports.
SENDER B: And doy you have hat? I wear my everywhere and everytime!
SENDER A: I have a few, that is a great idea!
SENDER B: Let`s take our hats and have a walk!
SENDER A: I love the smell of the trees and it is a beautiful day
SENDER B: Yes, sounds great! What is your job?
SENDER A: I work at a bank, and i work at a local place.
Example Output: Bot
Example Input: SENDER A: Hey, how old are you?
SENDER B: I'M 27 40 45 zitah?
SENDER A: How old are you?
SENDER B: I'M 27 40 45 won'tice 20 he'sn't
SENDER A: Anyways I’m 13 years old
SENDER B: That'S great! i am glad you are not that!
SENDER A: What is your hobby?
SENDER B: I love to cook, and i love to cook.
SENDER A: What exactly do you like to cook?
SENDER B: I cook a lot of italian food.
SENDER A: I love italian food!
SENDER B: I love italian food!
SENDER A: Unlike you I can’t really cook :/ I’m in a soccer team though!
SENDER B: I love soccer. i play soccer.
SENDER A: Yeah me too! My older brother plays soccer as well, he inspired me actually
SENDER B: I play soccer with my brother.
SENDER A: My brother is too busy to play soccer with me :/
SENDER B: I play soccer with my brother.
SENDER A: mmhm
SENDER B: I am a vegetarian, i love it.
SENDER A: oh wow how long have you been a vegetarian?
SENDER B: I have a vegetarian diet.
SENDER A: Okay, how long?
SENDER B: I'M 45 9am 40 poyou 400 1st 30 i'ver 250
SENDER A: I’ve gotta go, I’m gonna miss my school bus
SENDER B: I hope you are not in school
SENDER A: Bye!
SENDER B: I am not sure what you are doing
Example Output: | Human
| 3 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Let me give you an example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
The answer to this example can be: Bot
Here is why: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
OK. solve this:
SENDER A: yo
SENDER B: Dude, I don't understand you. Please, use English. (¬‿¬)
SENDER A: yo
SENDER B: Dude, I don't understand you. Please, use English. (¬‿¬)
SENDER A: hello
SENDER B: Hi, how are you doing?
Answer: | Human | 8 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example input: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Example output: Bot
Example explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Q: SENDER A: Helliw
SENDER B: I am a little girl who loves to go to the park
A: | Human | 3 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Q: SENDER A: i am not a jack , i am more of a recovering alcoholic .
SENDER B: hey, how're you?
SENDER A: i am good , how are you doing ?
SENDER B: fine thanks, nothing much
SENDER A: do you have any hobbies ?
SENDER B: not really, i like dancing,
SENDER A: i love dancing too ! i am a huge fan of the beatles myself .
SENDER B: good, so what do you do for living? if you don't mind
SENDER A: i am a hairdresser , but i am not doing it .
SENDER B: why is that?
SENDER A: i am not sure what you mean .
SENDER B: you mean you're a hairdresser but do you work as a hairdresser or not?
SENDER A: i am not sure what that is . i am a hairdresser .
SENDER B: ok, sorry, so are you a women or men barber mostly/
SENDER A: i am a single woman . i am a woman .
SENDER B: ah ok, and where are you from?
SENDER A: i am from the united states , how about you ?
SENDER B: i am from poland
SENDER A: i am a cali . cali .
SENDER B: nice to meet you, i'm iza
SENDER A: i love to meet people who have a friend
SENDER B: i have a few friends, you?
SENDER A: i have a ton of siblings . i am a big fan of them .
SENDER B: do they live far from you?
SENDER A: yes , they are in texas . i have a hard time choosing between the money .
SENDER B: ok, i see, so thanks for conversation, need to go, have a good day
SENDER A: you too . thanks for the compliment .
A: Bot
****
Q: SENDER A: Hi
SENDER B: I love green, reminds me of summer time.
SENDER A: My favorite color is orange, reminds of the sun
SENDER B: I love red, reminds me of summer time.
SENDER A: Do you have a pet?
SENDER B: I do not. I have a dog.
SENDER A: I have a fish
SENDER B: I have a fish, but i am not sure if i could eat fish.
SENDER A: How old are you?
SENDER B: I am 22 and i am a big fan of the nfl
SENDER A: I just turned 18
SENDER B: I am still in school, but i am looking for a job.
SENDER A: Are you dating?
SENDER B: I am not. I am just a regular time mom.
A: Human
****
Q: SENDER A: Good to see you!
SENDER B: Nice to meet you
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
A: | Bot
****
| 4 | NIv2 | task855_conv_ai_2_classification | fs_opt |
TASK DEFINITION: Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
PROBLEM: SENDER A: Dr. Lahiri, I am loving Chinese food.
SENDER B: What is your favourite music band?
SENDER A: III need music and dancing and secondhand smoke.
SENDER B: I think you are very funny person
SENDER A: Now, Within You Without You was not a commercial song, by any means.
SENDER B: I don't understand
SENDER A: Dancing Shadow, Single Guy, Miracle Musical, I'm Hungry.
SENDER B: What do you want for lunch
SENDER A: Wow, Miss BicBic's cooking is really delicious. Oh no, I'll be stuffing myself a lot again tonight!
SENDER B: Where are u from
SENDER A: Look, I'm a part of Shirazi Studios, too.
SENDER B: Where are you from
SENDER A: I... I live in Tel Aviv.
SENDER B: How old are you
SENDER A: Do I look that old?
SENDER B: No I think you are young
SENDER A: In Lausanne. Are you still in school?
SENDER B: Yes I'm a student of university
SENDER A: And, Professor, I'm taking your class again next semester.
SENDER B: Thanks for everything
SENDER A: Min Ji, Mother will call you again after your class.
SOLUTION: Bot
PROBLEM: SENDER A: [Game Show Music Plays] And thank you, Mr. Announcer.
SENDER B: Hi! how are you doing? are you going to the big party on the hill?
SENDER A: The Apple Picking Festival is a go but Great Adventure only wants three nights instead of 1 0.
SENDER B: There's going to be a big party on the hill. I love to party but I'm not sure if I should go there. It is on the edge of the hill, and I am deadly afraid of heights!
SENDER A: CHEERING AND APPLAUSE I would, you know, coach you through a whole bunch of song selections, out of your comfort zone. So...
SENDER B: you mean I should stay at home and listen to the music you recommend? alone? that's no party at all! even together with my dog and cat
SENDER A: Mr. Shuster, I want to do more of an upbeat, glamorous song with a really good dance beat.
SENDER B: are you even listening to me? is music all you want to talk about?
SENDER A: III need music and dancing and secondhand smoke.
SENDER B: yup, so let's go to the party if you're not afraid of heights! you could cheer me up there
SENDER A: Good. 'Cause I'm looking for dancers who can dance for 15 seconds 'cause that's how long my song is.
SENDER B: that's a very short song. and dance
SENDER A: Uh, I... I just said that this song is... is so great. I love it.
SENDER B: whatever. where's your song. let me hear it
SENDER A: (BOTH SINGING) You believe in rock 'n' roll?
SENDER B: hahaha isn't it dead?
SENDER A: So I... I thought since it's your song...
SENDER B: that should be enough I suppose...
SENDER A: So I... I thought since it's your song...
SENDER B: yup
SENDER A: DANCE MUSIC THUMPS Weirdest place you've ever done it.
SOLUTION: Bot
PROBLEM: SENDER A: hi
SENDER B: I like to drink a lot
SENDER A: me too. but they don't let us drink here in the juvenile detention center
SENDER B: Lol yea i think we might have to meet up and get to do that
SENDER A: they'll release me in about a month, we could drink to that
SENDER B: They do not have to do that. I am sure you are a good person.
SENDER A: nope. i was shoplifting. stole a bottle of bourbon
SENDER B: Oh, that is quite interesting.
SENDER A: not really. i don't like it here. hope my mom will come tomorrow to visit me and bring me something to smoke at least
SENDER B: That would be great. I will bring my dog to the park
SENDER A: well, good for you. you could come to visit me with your dog
SENDER B: We would love that. I am a very experienced person so i have to go.
SENDER A: where?
SENDER B: I am in texas.
SENDER A: what do you do there?
SENDER B: I am a stay at home mom.
SENDER A: you have many children?
SENDER B: No, i am not. I am only child.
SENDER A: how many children are you a mom to?
SENDER B: I am not sure, i am only 5
SENDER A: this is becoming weird
SOLUTION: | Human
| 8 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Q: SENDER A: Hello
SENDER B: I am not sure. I am more of a movie buff.
SENDER A: Ok
SENDER B: I am a huge gamer
SENDER A: K
SENDER B: I am in college.
SENDER A: K
SENDER B: I am a stay at home mom.
SENDER A: K
SENDER B: I am sure you are a little person. I am sure you are a great person.
SENDER A: K
SENDER B: I have a cat named radar
SENDER A: K
SENDER B: I am sure you are a little person. I am sure you are a great person.
SENDER A: K
SENDER B: I am sure you are a little person. I am sure you are a great person.
SENDER A: K
SENDER B: I am a stay at home mom.
SENDER A: K
SENDER A: K
SENDER A: J
SENDER A: Hello
SENDER A: Hey
A: Human
****
Q: SENDER A: hello
SENDER B: Hi, how are you doing?
SENDER A: i'm very excited and scared. i am expecting triplets in three months.
SENDER B: That'S a lot of kids. i have a 1 3 4 9
SENDER A: is that the age of your four kids? i have two step children already
SENDER B: No, i have two kids.
SENDER A: i like children. that's why i work at a school
SENDER B: I like to teach kids.
SENDER A: you're also a teacher? what do you teach them?
SENDER B: I teach math and teach them
SENDER A: I teach chemistry
SENDER B: I teach math and math
SENDER A: do you live in a house or apartment? i just moved into my new house for the first time
SENDER B: I live in the us, but i live in the city.
SENDER A: so it's an apartment building?
SENDER B: I have a garden in the city.
SENDER A: with fruit trees? what kind of trees are there?
SENDER B: 3000
A: Human
****
Q: SENDER A: Yo!
SENDER B: Hi!
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
A: | Bot
****
| 4 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Given the task definition, example input & output, solve the new input case.
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Output: Bot
This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
New input case for you: SENDER A: hi! how are you?
SENDER B: I'M doing great! how are you?
SENDER A: i'm great too. a little bored though
SENDER B: What do you do for a living?
SENDER A: i'm a joutnalist. and you?
SENDER B: I'M 34, i'm a very busy person.
SENDER A: where do you work?
SENDER B: I work at a bank.
SENDER A: wow, that's nice. do you like it?
SENDER B: Yes, i do. i love it.
SENDER A: happy for you. what are you doing now?
SENDER B: I am going to be a vet.
SENDER A: wait, you said you're working in a bank
SENDER B: I am a banker, but i am not very good at it.
SENDER A: why is that? have you considered learning?
SENDER B: I am not sure, i am not sure what you can tell me about yourself
SENDER A: well, i have two dogs
SENDER B: That'S cool. i have a dog.
SENDER A: what's its name?
SENDER B: I am not sure, i am a female.
SENDER A: funny
SENDER B: I like to laugh.
SENDER A: why do you want to be a vet?
SENDER B: I am a vet. i want to be a vet
SENDER A: no, you're a banker
SENDER B: That'S cool. i am a banker.
SENDER A: you are bizzare
SENDER B: I am not. i am a fan of bubblegum.
Output: | Human | 1 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Example solution: Bot
Example explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Problem: SENDER A: i am not very healthy . i work at a food store .
SENDER B: what kind of food do you sell?
SENDER A: i love all kinds of food . i am a vegetarian .
SENDER B: cool! I'm vegetarian too! that's why i decided to walk dogs for a living instead of eating them
SENDER A: i have a secret for gordan ramsey but i did not know it .
SENDER B: who is gordan ramsey and why you keep secrets from him?
SENDER A: he is a famous star wars fan .
SENDER B: he should be an avid kayaker fan just like me
SENDER A: i would say that would be fun to do .
SENDER B: i enjoy reading journals and guides related to psychology and it's even funnier to do! actually i'm lying just for instruction
SENDER A: you are funny , i read a lot about books . i am a huge fan of all types of books .
SENDER B: thats fine. whats your favorite books?
SENDER A: i like to read , i have been to new orleans .
SENDER B: are there book readers in orleans too?
SENDER A: yes , i have . i have read many books about it .
SENDER B: I can do a cell phone out of books because my parents taught me survival skills
SENDER A: we can not have the answer to them .
| Solution: Bot | 5 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Input: Consider Input: SENDER A: That'S cool i work at the local hospital
SENDER B: Hello how are you?
SENDER A: I am doing great, how are you?
SENDER B: Me too. How old are you?
SENDER A: I'M 6 feet tall
SENDER B: You are really tall. And what is your age?
SENDER A: 6 feet tall tall tall tall, i am 6 feet tall
SENDER B: You are not very smart aren't you?
SENDER A: Yes, i am. i am.
SENDER B: Who is your favourite singer?
SENDER A: I like country music, but i like all kinds
SENDER B: I like Taylor Swift
SENDER A: He'S won't say he is the name
SENDER B: I work at a school
SENDER A: Cool. i am a student.
SENDER B: What do you study?
SENDER A: I am studying to be an acara.
SENDER B: what is acara?
SENDER A: I like acara and acara.
SENDER B: and what else?
SENDER A: I like to listen to music
SENDER B: do you like jazz?
SENDER A: I do, but i like listening to music
SENDER B: i'm getting married in the spring.
SENDER A: Nice, i'm going to spring soon to be spring soon
Output: Bot
Input: Consider Input: SENDER A: hi!
SENDER B: Hi! how are you doing?
SENDER A: i'm okay. how are you?
SENDER B: I'M doing great! just got done with a new job.
SENDER A: congratulations! what job is that?
SENDER B: I work at a department
SENDER A: at which department?
SENDER B: I work at a department at the department
SENDER A: that's confusing. what do you do for a living?
SENDER B: I work at a local newspaper.
SENDER A: are you a journalist?
SENDER B: No, i am not. i am a writer.
SENDER A: well, a writer in a newspaper is a journalist, aren't they?
SENDER B: I have a book club.
SENDER A: that's cool. is it popular?
SENDER B: Yes, it is. i love it.
SENDER A: do you have any other hobbies?
SENDER B: I like to read and play with my friends
SENDER A: play what?
SENDER B: I play the guitar in a band.
SENDER A: you are so interesting. and i love mountain biking
SENDER B: I love hiking. i hike and hike.
SENDER A: i think hiking is boring
SENDER B: I like hiking. i like to hike.
SENDER A: i get that
SENDER B: I have a lot of money to do.
SENDER A: are you rich?
SENDER B: Yes, i am. i am 45 hasn't years old.
SENDER A: and i'm 26
SENDER B: I am a professional eater
SENDER A: i don't get it. are you a foog blogger?
SENDER B: I am not. i am not very good at it.
SENDER A: are you a food blogger?
SENDER B: Yes, i am. i am vegan.
SENDER A: why? you don't like meet?
SENDER B: I am not sure. i am going to be a professional.
SENDER A: a professional in what?
SENDER B: I work at a local newspaper.
SENDER A: you told that already
SENDER B: I am not sure what you mean
SENDER A: me too lol
SENDER B: I love to go to the park
SENDER A: and i love mountain biking
SENDER B: I like hiking too. i hike too.
SENDER A: that's boring
SENDER B: Yeah it is. i like it.
SENDER A: i gotta go
SENDER B: I love to ride my bike
SENDER A: bye
SENDER B: I am not sure what you are doing
Output: Human
Input: Consider Input: SENDER A: i am not very healthy . i work at a food store .
SENDER B: what kind of food do you sell?
SENDER A: i love all kinds of food . i am a vegetarian .
SENDER B: cool! I'm vegetarian too! that's why i decided to walk dogs for a living instead of eating them
SENDER A: i have a secret for gordan ramsey but i did not know it .
SENDER B: who is gordan ramsey and why you keep secrets from him?
SENDER A: he is a famous star wars fan .
SENDER B: he should be an avid kayaker fan just like me
SENDER A: i would say that would be fun to do .
SENDER B: i enjoy reading journals and guides related to psychology and it's even funnier to do! actually i'm lying just for instruction
SENDER A: you are funny , i read a lot about books . i am a huge fan of all types of books .
SENDER B: thats fine. whats your favorite books?
SENDER A: i like to read , i have been to new orleans .
SENDER B: are there book readers in orleans too?
SENDER A: yes , i have . i have read many books about it .
SENDER B: I can do a cell phone out of books because my parents taught me survival skills
SENDER A: we can not have the answer to them .
| Output: Bot
| 2 | NIv2 | task855_conv_ai_2_classification | fs_opt |
instruction:
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
question:
SENDER A: What?
SENDER B: I am a full time student
SENDER A: What?
SENDER B: I am a full time student
SENDER A: What universe you graduate?
SENDER B: I am studying to be a scientist.
SENDER A: Where youvwork?
SENDER B: I live in the north of the north of the coast of the coast.
SENDER A: Where you work?
SENDER B: I work at a gas station
SENDER A: How old are you?
SENDER B: I'M 27 40 45 won'tice 20 he'sn't
SENDER A: What is your education?
SENDER B: I am studying to be a teacher
answer:
Human
question:
SENDER A: What are you going to school for?
SENDER B: I'm going to build a robot!
SENDER A: That sounds exciting...............................................
SENDER B: Yes I know. Are you a robot?
SENDER A: No i am not sure i do not want to be a superhero
SENDER B: Do you watch youtube?
SENDER A: Yes, i do, i like youtube
SENDER B: Do you watch streams?
SENDER A: Yes, i do, i love to watch tv
SENDER B: What do you like to watch on youtube?
SENDER A: I like youtube videos, and i watch youtube.
SENDER B: What kind of videos do you like to watch?
SENDER A: I like youtube videos, what about you?
SENDER B: So do I. I like youtube.
SENDER A: I like youtube too..............................................
SENDER B: i am a recluse.
SENDER A: I am sorry to hear that.
SENDER B: I don't like people.
SENDER A: I like to talk to people
SENDER B: I don't talk to people very often.
SENDER A: I like to talk to people
SENDER B: What do you like to talk about?
SENDER A: I like to talk about friends
SENDER B: Do you have a lot of friends?
SENDER A: Yes, we have a lot of friends. we have a friends.
SENDER B: Who is your best friend?
SENDER A: I like all kinds, and i like all kinds of music
answer:
Bot
question:
SENDER A: Good to see you!
SENDER B: Good to see you too
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: How are you doing?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: Hello??
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: Are you okay?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
answer:
| Bot
| 9 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Teacher: Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Teacher: Now, understand the problem? If you are still confused, see the following example:
SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Reason: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Now, solve this instance: SENDER A: Ok
SENDER B: I am not sure what you do for a living?
SENDER A: What is your name?
SENDER B: He'S 59th
Student: | Human | 2 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Detailed Instructions: Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
See one example below:
Problem: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Problem: SENDER A: After Coming From Reality, did you wanna continue making albums?
SENDER B: Yes
SENDER A: I'm leading with it, to put the audience at their ease.
Solution: | Bot | 4 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Teacher: Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Teacher: Now, understand the problem? If you are still confused, see the following example:
SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Reason: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Now, solve this instance: SENDER A: i am not a jack , i am more of a recovering alcoholic .
SENDER B: hey, how're you?
SENDER A: i am good , how are you doing ?
SENDER B: fine thanks, nothing much
SENDER A: do you have any hobbies ?
SENDER B: not really, i like dancing,
SENDER A: i love dancing too ! i am a huge fan of the beatles myself .
SENDER B: good, so what do you do for living? if you don't mind
SENDER A: i am a hairdresser , but i am not doing it .
SENDER B: why is that?
SENDER A: i am not sure what you mean .
SENDER B: you mean you're a hairdresser but do you work as a hairdresser or not?
SENDER A: i am not sure what that is . i am a hairdresser .
SENDER B: ok, sorry, so are you a women or men barber mostly/
SENDER A: i am a single woman . i am a woman .
SENDER B: ah ok, and where are you from?
SENDER A: i am from the united states , how about you ?
SENDER B: i am from poland
SENDER A: i am a cali . cali .
SENDER B: nice to meet you, i'm iza
SENDER A: i love to meet people who have a friend
SENDER B: i have a few friends, you?
SENDER A: i have a ton of siblings . i am a big fan of them .
SENDER B: do they live far from you?
SENDER A: yes , they are in texas . i have a hard time choosing between the money .
SENDER B: ok, i see, so thanks for conversation, need to go, have a good day
SENDER A: you too . thanks for the compliment .
Student: | Bot | 2 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Detailed Instructions: Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
See one example below:
Problem: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Problem: SENDER A: I like nascar too, i like nascar
SENDER B: Hey, i am Sammy, i work at a bar at night to pay for college
SENDER A: Nice! i am in college, but i am not good at it.
SENDER B: i'm in college too, to become a nurse
SENDER A: That'S a good job. i am a nurse.
SENDER B: WOW. What else do you like? Cars?
SENDER A: I like to watch cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars,
SENDER B: it is many cars
SENDER A: I like cars, like cars
SENDER B: i see
SENDER A: I have a dog named bella
SENDER B: oo, that is sweet. how old she is
SENDER A: She'S 18 years of age.
SENDER B: granny)
SENDER A: Hello there! how are you?
SENDER B: Fine, i've blonde hair now
SENDER A: I am blonde and love it
SENDER B: and blue eyes?
SENDER A: I like blue, but i like blue
SENDER B: i like nascar)
SENDER A: I like to play with my friends
SENDER B: i have no time for friends, are you married
SENDER A: Yes, i have two girlfriends
SENDER B: too many)
SENDER A: I like to make sure.
SENDER B: 🙈
Solution: | Bot | 4 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Q: SENDER A: i am not sure what that is . i am not a very experienced person .
SENDER B: How are you?
SENDER A: i am not sure . i am a very experienced person .
SENDER B: I got it. How old are you?
SENDER A: no i have not tried it . i do not think i could do that
SENDER B: You are so insecure
SENDER A: i think i am . i am sure that is true .
SENDER B: You should be a little more confident
SENDER A: i think so . i am sure you are very attractive .
SENDER B: Thanks. How old are you?
SENDER A: i am very short . i have to go to work soon .
SENDER B: What is your job?
SENDER A: i am not sure . i am a very experienced person .
SENDER B: Do you have a job?
SENDER A: i am not sure , i am a very experienced person .
A: Bot
****
Q: SENDER A: I like nascar too, i like nascar
SENDER B: Hey, i am Sammy, i work at a bar at night to pay for college
SENDER A: Nice! i am in college, but i am not good at it.
SENDER B: i'm in college too, to become a nurse
SENDER A: That'S a good job. i am a nurse.
SENDER B: WOW. What else do you like? Cars?
SENDER A: I like to watch cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars,
SENDER B: it is many cars
SENDER A: I like cars, like cars
SENDER B: i see
SENDER A: I have a dog named bella
SENDER B: oo, that is sweet. how old she is
SENDER A: She'S 18 years of age.
SENDER B: granny)
SENDER A: Hello there! how are you?
SENDER B: Fine, i've blonde hair now
SENDER A: I am blonde and love it
SENDER B: and blue eyes?
SENDER A: I like blue, but i like blue
SENDER B: i like nascar)
SENDER A: I like to play with my friends
SENDER B: i have no time for friends, are you married
SENDER A: Yes, i have two girlfriends
SENDER B: too many)
SENDER A: I like to make sure.
SENDER B: 🙈
A: Bot
****
Q: SENDER A: I like nascar too, i like nascar
SENDER B: Hey, i am Sammy, i work at a bar at night to pay for college
SENDER A: Nice! i am in college, but i am not good at it.
SENDER B: i'm in college too, to become a nurse
SENDER A: That'S a good job. i am a nurse.
SENDER B: WOW. What else do you like? Cars?
SENDER A: I like to watch cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars, cars,
SENDER B: it is many cars
SENDER A: I like cars, like cars
SENDER B: i see
SENDER A: I have a dog named bella
SENDER B: oo, that is sweet. how old she is
SENDER A: She'S 18 years of age.
SENDER B: granny)
SENDER A: Hello there! how are you?
SENDER B: Fine, i've blonde hair now
SENDER A: I am blonde and love it
SENDER B: and blue eyes?
SENDER A: I like blue, but i like blue
SENDER B: i like nascar)
SENDER A: I like to play with my friends
SENDER B: i have no time for friends, are you married
SENDER A: Yes, i have two girlfriends
SENDER B: too many)
SENDER A: I like to make sure.
SENDER B: 🙈
A: | Bot
****
| 4 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Given the task definition, example input & output, solve the new input case.
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Output: Bot
This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
New input case for you: SENDER A: Alright mate?
SENDER B: yup. I'm cooking dinner for my friends
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
Output: | Bot | 1 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
--------
Question: SENDER A: In Las Vegas, you're an afterthought, you're an old so while we could go gambling, we could go hookering, we could get drunk, we could go to the convention...
SENDER B: hello
SENDER A: Hey, Mr. San Francisco, I'd like you to meet...
SENDER B: i love britney spears and you?
SENDER A: The New York City Ballet Gala, duh.
SENDER B: i'm on a diet now, have you ever followed any diet?
SENDER A: Thanks, Zeke! No way. My sparkle jelly stays on my wrist until I die, and then probably even after that.
SENDER B: lol, sounds funny
SENDER A: Big Buddha Cheese for muscle spasms.
SENDER B: do you like spending your (or not so) money?
SENDER A: Look, II have got... 25 bucks for food.
SENDER B: could you give them to me?
SENDER A: This Friday. How much did you sell 'em?
SENDER B: a lot, lol
SENDER A: II'm not gonna sell them.
SENDER B: well, why?..
SENDER A: But I'll pay you like a regular customer.
SENDER B: oh, thanks
SENDER A: I... I don't mean selling programmes or finding prizes for the tombola.
SENDER B: l should go, sorry
SENDER A: I... I'd rather stay here, if you don't mind.
SENDER B: no, i don't, stay here, buddy
SENDER A: I'll go on alone.
SENDER B: bye
SENDER A: Hi, Les, I'll be right with you.
Answer: Bot
Question: SENDER A: i am not a soda fan . i eat too much .
SENDER B: Hello!
SENDER A: i am a very experienced person , so i know it is a definite no !
SENDER B: Do yoy have problem with alcohol?
SENDER A: i have no idea what that is
SENDER B: But I`m recovering now. You don`t drink at all, do you?
SENDER A: i have no idea what that is
SENDER B: Where are you from?
SENDER A: i am not sure . i am a very experienced person .
SENDER B: Wonderful! Do you take any meditation from time to time?
SENDER A: i have tried to try it . i have to try it .
SENDER B: Oh, you should try it more and more! I love it!
SENDER A: i have not tried that . i am sure it will .
SENDER B: Now I live in the city and I have a lot of time to spend with meditation.
SENDER A: i have never heard of that .
SENDER B: Now you do. Where are you from?
SENDER A: i have never tried it , but i will try it .
Answer: Bot
Question: SENDER A: нщ
SENDER B: Dude, I don't understand you. Please, use English. (¬‿¬)
SENDER A: yo
Answer: | Human
| 7 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Teacher: Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Teacher: Now, understand the problem? If you are still confused, see the following example:
SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Reason: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Now, solve this instance: SENDER A: hello
SENDER B: I am a huge fan of the outdoors. I am a bit of a loner.
Student: | Human | 2 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
--------
Question: SENDER A: I am the youngest, but my brother is the youngest in my brother
SENDER B: so you are the youngest... sister?
SENDER A: Yes, i'm. my sister is in my family
SENDER B: how many brothers and sisters do you have?
SENDER A: I have three sisters and they're 30 pairs of 12 children
SENDER B: what do you do in life? my wife is a supermodel from Italy
SENDER A: I work at a hospital in the basement, there's no italian
SENDER B: what? in the basement? you mean the morgue???
SENDER A: I am in the morgue, i am in a morgue apartment.
SENDER B: that must be very scary! I work with bees but it/s better than working with dead people
SENDER A: I bet they are, i am allergic to bees
SENDER B: I'm allergic to death hahaha
SENDER A: I am allergic to peanuts, so i have to it in the whole life
SENDER B: which country do you live in?
SENDER A: I live in the south, florida
SENDER B: must be hot there this time of year
SENDER A: Yeah it is a good one
Answer: Bot
Question: SENDER A: Hey I’m 46-year-old single dad
SENDER B: I am sorry to hear that.
SENDER A: My ex cheated me
SENDER B: I am sorry to hear that.
SENDER A: Left me for all
SENDER B: I am sorry to hear that.
SENDER A: Prefer being inside
SENDER B: I like to eat. i eat a lot of indian food.
SENDER A: Cool where do you live I live in California
SENDER B: I live in california, where are you from?
SENDER A: Me too
SENDER B: I love to go to the park
SENDER A: You know my job
SENDER B: I am a cashier at a gas station
SENDER A: I’m recording engineer I have my own
SENDER B: I bet you do not have a lot of money.
SENDER A: Yeah kind of
SENDER B: I like to go to the gym
SENDER A: LOL
SENDER B: I am not sure what you mean
SENDER A: Laugh out loud
SENDER B: I like to laugh. i like to laugh.
SENDER A: Cool you know what I need to leave now
SENDER B: I am a professional athlete, but i do not know what you mean
SENDER A: Goodbye
SENDER B: I am sorry to hear that.
SENDER A: Bye
SENDER B: I am not sure what you are doing
SENDER A: My kids are calling me I need to go
SENDER B: I have a 1 daughter.
SENDER A: Bye
SENDER B: I am not sure what you are doing
SENDER A: I need to go to the hospital
SENDER B: I hope you are a doctor.
SENDER A: No I’m sick
SENDER B: I am sorry to hear that.
SENDER A: No I need to leave goodbye
SENDER B: I am sorry to hear that.
SENDER A: See you
SENDER B: I do not know. i have a few of them.
SENDER A: By boat
SENDER B: I love fishing. i have a boat that i have a boat.
SENDER A: Me too
SENDER B: I love to go to the park
SENDER A: I need to go
SENDER B: I will be going to be a doctor.
SENDER A: I hope
SENDER B: I hope you are not.
SENDER A: Goodbye
SENDER B: I am sorry to hear that.
SENDER A: Goodbye
SENDER B: I am sorry to hear that.
Answer: Human
Question: SENDER A: Alright mate?
SENDER B: I'm alright. How are you?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: Hey... Let's talk
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: Why are you quiet?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: Say something
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
Answer: | Bot
| 7 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Part 1. Definition
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Part 2. Example
SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Answer: Bot
Explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Part 3. Exercise
SENDER A: I speak spanish, spanish, spanish, spanish, spanish, spanish, spanish
SENDER B: I don't. do you speak english?
SENDER A: I speak english and english.
SENDER B: fnd spanish )) you like spanish food? I love shawarma
SENDER A: I love pizza!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
SENDER B: that's Itallian )))
SENDER A: Hello there! how are you?
SENDER B: hey. I'm good. you?
SENDER A: I am good, just finished watching some tv
SENDER B: what else do you do?
SENDER A: I like to write stories and write
SENDER B: stories about what?
SENDER A: I read about stories about stories about stories about stories about stories about stories about stories.
SENDER B: you are funny. what's your name?
SENDER A: Yes, i am. i am.
SENDER B: do you go to school?
SENDER A: Yes, i do, i am currently majoring in college
SENDER B: I never even graduated from school, I went to the army
SENDER A: I did not know what that means
Answer: | Bot | 7 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Example solution: Bot
Example explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Problem: SENDER A: Ok
SENDER B: I am not sure what you do for a living?
| Solution: Human | 5 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
--------
Question: SENDER A: Hellow
SENDER B: Dude, I don't understand you. Please, use English. (¬‿¬)
Answer: Human
Question: SENDER A: Hi
SENDER B: I am a little girl, but i am not a fan of any animals.
SENDER A: How old are you?
SENDER B: I am 22 and i am a big fan of the nfl
SENDER A: Why you decade to change male?
SENDER B: I am not sure about that. I am not sure.
SENDER A: What is your male
SENDER B: I am not sure what that means.
SENDER A: What is your name?
SENDER B: I am steve. I am not sure what you mean.
Answer: Human
Question: SENDER A: hello
SENDER B: Hi, how are you doing?
SENDER A: i'm good. just finished reading a book
SENDER B: That'S cool, what do you like to read?
SENDER A: oh i read everything. i read at least 20 books a year
SENDER B: That'S cool. i read a lot of books
SENDER A: what is your favorite book?
SENDER B: I like reading books, 1984
SENDER A: that's a good one from George Orwell. so you like antiuthopia?
SENDER B: I do, i like to read.
SENDER A: what else do you do?
SENDER B: I work at a local animal shelter.
SENDER A: do they kill animals that they couldn't give away for adoption?
SENDER B: They'Re 42 3rd 3d
SENDER A: is that an address?
SENDER B: Yes, it is. i am not sure it is.
SENDER A: neither am i
SENDER B: I am a vegetarian, i am a vegetarian.
SENDER A: that makes sense. you like animals, why would you eat them
SENDER B: I think i would be happy to be honest.
Answer: | Human
| 7 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Detailed Instructions: Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
See one example below:
Problem: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Problem: SENDER A: hello
SENDER B: Hi, how are you doing?
SENDER A: pretty good. it's nice here at the country farm
SENDER B: I love the farm!
SENDER A: yeah? well come on over! i'll introduce you to my pet cow. her name is Gladis
SENDER B: I have a cat named cow.
SENDER A: hahahahaha you're kidding, right?
SENDER B: I think it is. i am a very good person.
SENDER A: you seem like a good person to me. what do you do in life?
SENDER B: I work at a bank, and i love to travel.
SENDER A: where did you traver already?
SENDER B: I live in the midwest in the midwest
SENDER A: so do you travel the usa or go to other countries?
SENDER B: I live in canada. i travel a lot.
Solution: | Human | 4 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example input: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Example output: Bot
Example explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Q: SENDER A: hello
SENDER B: Hi, how are you doing?
SENDER A: i'm good. just finished reading a book
SENDER B: That'S cool, what do you like to read?
SENDER A: oh i read everything. i read at least 20 books a year
SENDER B: That'S cool. i read a lot of books
SENDER A: what is your favorite book?
SENDER B: I like reading books, 1984
SENDER A: that's a good one from George Orwell. so you like antiuthopia?
SENDER B: I do, i like to read.
SENDER A: what else do you do?
SENDER B: I work at a local animal shelter.
SENDER A: do they kill animals that they couldn't give away for adoption?
SENDER B: They'Re 42 3rd 3d
SENDER A: is that an address?
SENDER B: Yes, it is. i am not sure it is.
SENDER A: neither am i
SENDER B: I am a vegetarian, i am a vegetarian.
SENDER A: that makes sense. you like animals, why would you eat them
SENDER B: I think i would be happy to be honest.
A: | Human | 3 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Ex Input:
SENDER A: Hallo! What is your name?
SENDER B: Hello jake, i am doing well. How are you?
SENDER A: I' m fine! What are you doing now?
SENDER B: I am doing well. I am just hanging out with my dog.
SENDER A: Oh! Have you a dog?
SENDER B: I am a cat lady. I have a dog named wiggles.
SENDER A: Oh! Have you a dog?
SENDER B: I would love to go there.
SENDER A: Oh! Has you a dog?
SENDER B: I would love to go there.
SENDER A: Oh! Have you a dog?
SENDER B: I love to cook, but i am not good at it.
SENDER A: What do you cook?
SENDER B: I eat a lot of meat, but i am not good at it.
SENDER A: What do you cook?
SENDER B: I am not into fashion. I am a very experienced writer.
SENDER A: What do you write?
SENDER B: I am a writer, i love to read and write poetry. You?
SENDER A: I like to read eanglish poetry!
Ex Output:
Human
Ex Input:
SENDER A: hello
SENDER B: Hi, how are you doing?
SENDER A: ok
SENDER B: I am not sure what you do for a living?
SENDER A: ok
SENDER B: I am not sure what you do for a living?
Ex Output:
Human
Ex Input:
SENDER A: Hi
SENDER B: Hi, how are you doing?
SENDER A: How are you
SENDER B: I'M doing great, how about you?
SENDER A: Do you speak german?
SENDER B: No, i speak german, but i speak french and english.
SENDER A: Parle france
SENDER B: Dude, I don't understand you. Please, use English. (¬‿¬)
SENDER A: Common sava?
SENDER B: Yes, i am. i am a very good person.
SENDER A: Do you speak a french?
SENDER B: No, i speak french and english.
Ex Output:
| Human
| 1 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Ex Input:
SENDER A: Hi
SENDER B: My mom is a teacher, but i am not sure what that means.
SENDER A: Whatvis your male
SENDER B: I am not sure. I am not sure what that means.
SENDER A: ?
SENDER B: I am not sure what that means.
SENDER A: What isbyour male?
SENDER B: I like to eat fish, i am not a fan of it.
SENDER A: You are steve?
SENDER B: Yes i am. I am a very good person.
SENDER A: Ok, bob. What is your name?
SENDER B: I am steve. I am not sure what that is.
SENDER A: Your nameb is bob?
SENDER B: No i am not sure what that means.
SENDER A: Your name is bob?
SENDER B: I am steve. I am a cat named after my favorite thing to do.
SENDER A: What is your male?
SENDER B: I am a male, and i am a real estate agent
SENDER A: Prove what you are not bot
SENDER A: Bob?
SENDER B: I am not a big fan of that. I am more into comedies.
SENDER A: Why you parents call your Bob?
SENDER B: I am not very artistic. I am more of a quiet guy.
SENDER A: What is your name?
SENDER B: I am steve. I am not sure what you mean.
SENDER A: Are you crazy?
SENDER B: I am not. I am a stay at home mom.
SENDER A: Ehy?
SENDER B: Yes, i am a very experienced person.
SENDER A: Why?
SENDER B: I am not sure what you mean.
SENDER A: How old are you?
SENDER B: I am 22 and i am a big fan of the nfl
SENDER A: You are 23?
SENDER B: Yes i am. I am a huge fan of all animals.
SENDER A: You are 13?
SENDER B: I am. I am a huge fan of all animals.
SENDER A: Aou are female?
SENDER B: I think i am not sure. I am a very experienced person.
SENDER A: You are animal?
SENDER B: I am not sure what i want to do with them
SENDER A: How old are you?
SENDER B: I love to go camping
Ex Output:
Human
Ex Input:
SENDER A: hello
SENDER B: I am a huge fan of the outdoors. I am a bit of a loner.
Ex Output:
Human
Ex Input:
SENDER A: I am a professional wrestler in wwe.What is your job?
SENDER B: I work at a local.
SENDER A: I'm also volunteer in a homeless shelter.Do you know any people around who is homeless?I can help them.
SENDER B: I am a vet. i am a vet.
SENDER A: Good i have a dog and i think it feels sick, can you help it?
SENDER B: I have a dog named tiki
SENDER A: My dog's name is allen.
SENDER B: I am from the south of the desert.
SENDER A: Good, im from there too!What is your favourite food in there?
SENDER B: I love pizza!
SENDER A: Me too!What do you love about pizza
SENDER B: I love pizza!
SENDER A: Me too.But im a chicken person myself.
SENDER B: I love chicken. i eat chicken.
SENDER A: I love it too!
SENDER B: I love it! i love it!
SENDER A: What is your hobby
SENDER B: I love to cook, and you?
SENDER A: I love to cook too.What is your favourite meal for cooking
SENDER B: I love cooking, and cooking.
SENDER A: Do you know the recipe of carrot cake?
SENDER B: I do not know. i have a pumpkin.
Ex Output:
| Human
| 1 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
[EX Q]: SENDER A: Hi I’m Jonathan
SENDER B: Hi, how are you doing?
SENDER A: I’m good you
SENDER B: I am doing well, how about you?
SENDER A: What’s your favorite activity
SENDER B: I love to go hiking
SENDER A: Mines is rock climbing
SENDER B: That'S cool. i like to hike.
SENDER A: Good
SENDER B: I am doing well, how about you?
SENDER A: What’s your job
SENDER B: 1984, and i work at a local newspaper.
SENDER A: I’m a mechanical engineer
SENDER B: That'S cool. i work at a computer company.
SENDER A: I have lab named black troop
SENDER B: That'S cool. i have a lab.
SENDER A: Where
SENDER B: I live in the midwest in the midwest
SENDER A: Where are you from
SENDER B: I live in the usa
SENDER A: Cool
SENDER B: Yeah, i love it.
SENDER A: You know what I need to go now I am going home see you later ok?
SENDER B: I am going to be a vet.
SENDER A: Nice meeting you
SENDER B: I am going to be a little tired
SENDER A: Bye
SENDER B: I am not sure what you are doing
SENDER A: I need to leave now
SENDER B: I am sure you can find a good idea.
[EX A]: Human
[EX Q]: SENDER A: Ok
SENDER B: I am not sure what you do for a living?
SENDER A: What is your name?
SENDER B: He'S 59th
[EX A]: Human
[EX Q]: SENDER A: Whazzup?
SENDER B: Are you here?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
[EX A]: | Bot
| 6 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Part 1. Definition
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Part 2. Example
SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Answer: Bot
Explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Part 3. Exercise
SENDER A: Lol
SENDER B: I am not sure what you mean
SENDER A: Fucking bot
SENDER B: I am a huge fan of it.
SENDER A: What is your name?
SENDER B: He'S 59th
SENDER A: How old are you?
SENDER B: I'M 27 40 45 won'tice 20 he'sn't
Answer: | Human | 7 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
[EX Q]: SENDER A: Hellow
SENDER B: Dude, I don't understand you. Please, use English. (¬‿¬)
SENDER A: Do you speak french?
SENDER B: I speak french and french and french
SENDER A: Do you speak german?
SENDER B: No, i speak german, but i speak french and english.
[EX A]: Human
[EX Q]: SENDER A: Hello there!
SENDER B: i am not sure . i am a very experienced person .
SENDER A: Do you like cars? I like beemers!
SENDER B: i did not , i was thinking about it .
SENDER A: There is no no think about! Get good education like me in yale, buy a bmw and you on a top of life!
SENDER B: i have heard of it . i am a pagan .
SENDER A: Pagan... I am Christian!
SENDER B: i have never heard of that , but it is really fun .
SENDER A: Not so fun, my parents will cut me off money flow if i go pagan
SENDER B: backstreet boys are cute . i am a huge fan of that .
[EX A]: Human
[EX Q]: SENDER A: Lol
SENDER B: I am not sure what you mean
SENDER A: Fucking bot
SENDER B: I am a huge fan of it.
SENDER A: What is your name?
SENDER B: He'S 59th
SENDER A: How old are you?
SENDER B: I'M 27 40 45 won'tice 20 he'sn't
[EX A]: | Human
| 6 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
SENDER A: How’s it going?
SENDER B: good
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: /test
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: /test
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
Bot
SENDER A: i am not a jack , i am more of a recovering alcoholic .
SENDER B: hey, how're you?
SENDER A: i am good , how are you doing ?
SENDER B: fine thanks, nothing much
SENDER A: do you have any hobbies ?
SENDER B: not really, i like dancing,
SENDER A: i love dancing too ! i am a huge fan of the beatles myself .
SENDER B: good, so what do you do for living? if you don't mind
SENDER A: i am a hairdresser , but i am not doing it .
SENDER B: why is that?
SENDER A: i am not sure what you mean .
SENDER B: you mean you're a hairdresser but do you work as a hairdresser or not?
SENDER A: i am not sure what that is . i am a hairdresser .
SENDER B: ok, sorry, so are you a women or men barber mostly/
SENDER A: i am a single woman . i am a woman .
SENDER B: ah ok, and where are you from?
SENDER A: i am from the united states , how about you ?
SENDER B: i am from poland
SENDER A: i am a cali . cali .
SENDER B: nice to meet you, i'm iza
SENDER A: i love to meet people who have a friend
SENDER B: i have a few friends, you?
SENDER A: i have a ton of siblings . i am a big fan of them .
SENDER B: do they live far from you?
SENDER A: yes , they are in texas . i have a hard time choosing between the money .
SENDER B: ok, i see, so thanks for conversation, need to go, have a good day
SENDER A: you too . thanks for the compliment .
Bot
SENDER A: Look, Bert. If I choose to live like a... like a church mouse, I have a reason.
SENDER B: Are you religious?
SENDER A: The Charismatic Catholics need an extra half hour this morning to learn a new hymn, so if you could just move along...
SENDER B: Don't you want to talk to me?
SENDER A: The Old Church, Amsterdam. You don't normally think of armpits in a church.
SENDER B: Armpits in church, what?
SENDER A: The Charismatic Catholics need an extra half hour this morning to learn a new hymn, so if you could just move along...
SENDER B: I got it already. Are you Catholic?
SENDER A: Oh, Jesus! I feel like a whore in a Catholic church!
SENDER B: Don't say that word! Are you sinner?
SENDER A: My Lord Jesus Christ, true God and man,... for being who You are, and because I love you above all things,...
SENDER B: Do you love me?
SENDER A: Why ? Because I need to sin.
SENDER B: I don't understand you
SENDER A: The Catholic Church should go to confession, not you.
SENDER B: Okay. Let's move on. How old are you?
SENDER A: Oh, Jesus! I feel like a whore in a Catholic church!
SENDER B: Dude. Stop it
SENDER A: Okay. God, I'm really kicking your ass.
SENDER B: You are so rude
SENDER A: No! You don't behave, you go in the bad girl closet.
SENDER B: I'm just a little girl
SENDER A: No! Sometimes I'm jealous of me.
SENDER B: Okay, bye
SENDER A: 'Cause Janny Boo is driving me nuts!
| Bot
| 0 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Ex Input:
SENDER A: кп
SENDER A: finished. Thank you for participation! Save somewhere your secret conversation ID.
SENDER B: i love green . my hair is green .
Ex Output:
Human
Ex Input:
SENDER A: What are you going to school for?
SENDER B: I'm going to build a robot!
SENDER A: That sounds exciting...............................................
SENDER B: Yes I know. Are you a robot?
SENDER A: No i am not sure i do not want to be a superhero
SENDER B: Do you watch youtube?
SENDER A: Yes, i do, i like youtube
SENDER B: Do you watch streams?
SENDER A: Yes, i do, i love to watch tv
SENDER B: What do you like to watch on youtube?
SENDER A: I like youtube videos, and i watch youtube.
SENDER B: What kind of videos do you like to watch?
SENDER A: I like youtube videos, what about you?
SENDER B: So do I. I like youtube.
SENDER A: I like youtube too..............................................
SENDER B: i am a recluse.
SENDER A: I am sorry to hear that.
SENDER B: I don't like people.
SENDER A: I like to talk to people
SENDER B: I don't talk to people very often.
SENDER A: I like to talk to people
SENDER B: What do you like to talk about?
SENDER A: I like to talk about friends
SENDER B: Do you have a lot of friends?
SENDER A: Yes, we have a lot of friends. we have a friends.
SENDER B: Who is your best friend?
SENDER A: I like all kinds, and i like all kinds of music
Ex Output:
Bot
Ex Input:
SENDER A: How’s it going?
SENDER B: Thanks, great. How are you?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: /test
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
Ex Output:
| Bot
| 1 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
One example is below.
Q: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
A: Bot
Rationale: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Q: SENDER A: Hellow
SENDER B: Dude, I don't understand you. Please, use English. (¬‿¬)
SENDER A: Do you speak french?
SENDER B: I speak french and french and french
SENDER A: Do you speak german?
SENDER B: No, i speak german, but i speak french and english.
A: | Human | 9 | NIv2 | task855_conv_ai_2_classification | fs_opt |
You will be given a definition of a task first, then an example. Follow the example to solve a new instance of the task.
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Why? This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
New input: SENDER A: hello
SENDER B: Hi, how are you doing?
SENDER A: ok
SENDER B: I am not sure what you do for a living?
SENDER A: ok
SENDER B: I am not sure what you do for a living?
Solution: | Human | 0 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Input: Consider Input: SENDER A: hello
SENDER B: Hi, how are you doing?
SENDER A: i'm very excited and scared. i am expecting triplets in three months.
SENDER B: That'S a lot of kids. i have a 1 3 4 9
SENDER A: is that the age of your four kids? i have two step children already
SENDER B: No, i have two kids.
SENDER A: i like children. that's why i work at a school
SENDER B: I like to teach kids.
SENDER A: you're also a teacher? what do you teach them?
SENDER B: I teach math and teach them
SENDER A: I teach chemistry
SENDER B: I teach math and math
SENDER A: do you live in a house or apartment? i just moved into my new house for the first time
SENDER B: I live in the us, but i live in the city.
SENDER A: so it's an apartment building?
SENDER B: I have a garden in the city.
SENDER A: with fruit trees? what kind of trees are there?
SENDER B: 3000
Output: Human
Input: Consider Input: SENDER A: Hallo! What is your name?
SENDER B: Hello jake, i am doing well. How are you?
SENDER A: I' m fine! What are you doing now?
SENDER B: I am doing well. I am just hanging out with my dog.
SENDER A: Oh! Have you a dog?
SENDER B: I am a cat lady. I have a dog named wiggles.
SENDER A: Oh! Have you a dog?
SENDER B: I would love to go there.
SENDER A: Oh! Has you a dog?
SENDER B: I would love to go there.
SENDER A: Oh! Have you a dog?
SENDER B: I love to cook, but i am not good at it.
SENDER A: What do you cook?
SENDER B: I eat a lot of meat, but i am not good at it.
SENDER A: What do you cook?
SENDER B: I am not into fashion. I am a very experienced writer.
SENDER A: What do you write?
SENDER B: I am a writer, i love to read and write poetry. You?
SENDER A: I like to read eanglish poetry!
Output: Human
Input: Consider Input: SENDER A: i am not sure what that means . i am more of a couch potato .
SENDER B: hello i just moved here from germany
SENDER A: are you japanese ? i am from the united kingdom .
| Output: Bot
| 2 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Teacher: Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Teacher: Now, understand the problem? If you are still confused, see the following example:
SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Reason: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Now, solve this instance: SENDER A: i am not a soda fan . i eat too much .
SENDER B: Hello!
SENDER A: i am a very experienced person , so i know it is a definite no !
SENDER B: Do yoy have problem with alcohol?
SENDER A: i have no idea what that is
SENDER B: But I`m recovering now. You don`t drink at all, do you?
SENDER A: i have no idea what that is
SENDER B: Where are you from?
SENDER A: i am not sure . i am a very experienced person .
SENDER B: Wonderful! Do you take any meditation from time to time?
SENDER A: i have tried to try it . i have to try it .
SENDER B: Oh, you should try it more and more! I love it!
SENDER A: i have not tried that . i am sure it will .
SENDER B: Now I live in the city and I have a lot of time to spend with meditation.
SENDER A: i have never heard of that .
SENDER B: Now you do. Where are you from?
SENDER A: i have never tried it , but i will try it .
Student: | Bot | 2 | NIv2 | task855_conv_ai_2_classification | fs_opt |
TASK DEFINITION: Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
PROBLEM: SENDER A: Hello, how are you doing?
SENDER B: I am doing well. How are you?
SENDER A: I’m doing okay I guess, just a little lonely
SENDER A: So tell me about yourself
SENDER B: I am a stay at home mom.
SENDER A: What do you mean exactly?
SENDER B: I am a woman, i am a woman, but i am not looking for a boyfriend.
SENDER A: I see
SENDER B: I am not a big fan of that. I am more into movies.
SENDER A: Do you have any superpowers?
SENDER B: No, i do not. I am more of a loner.
SENDER A: I do have some and I like making time stop
SENDER B: I spend too much time with my job at home.
SENDER A: A job? How old are you?
SENDER B: I am in my sixties and i am a retired actress
SENDER A: So what job do you do at home if you’re retired?
SENDER A: Hello?
SENDER B: Hello, i am doing well. How are you?
SENDER A: I’m fine thanks
SENDER A: Where do you live?
SENDER B: I live in california, near the beach. You?
SENDER A: I live on the cloud
SENDER A: Okay i’ve gotta go, see you
SOLUTION: Human
PROBLEM: SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Not good
SENDER B: I am sorry to hear that.
SENDER A: Ok
SENDER B: I am not sure what you do for a living?
SENDER A: I am a professional duck fucker
SENDER B: That'S cool. i like to play ball
SENDER A: Do you like to suck balls?
SENDER B: I do, but i do not like to bake.
SENDER A: So don't bake
SENDER B: I am a teacher, but i do not have a job
SENDER A: You are stupid teacher
SENDER B: I am not sure, i am 34, but i am not very good at it.
SENDER A: Are you not good at been 34? Are you kidding me?
SENDER B: No, i am not sure. i am not sure i am.
SENDER A: Ok
SENDER B: I am not sure what you do for a living?
SENDER A: I told you
SENDER B: I am sorry to hear that.
SENDER A: You are stupid
SENDER B: I am not sure what you mean.
SOLUTION: Human
PROBLEM: SENDER A: i am not sure what that is . i am more of a cat person .
SENDER B: Hey ConvAI, i am Rüstem. I am at junior grade, age is 17, wanted to chat with you a litte about my crush
SENDER B: Are you there _
SENDER A: i am a virtual . i am not sure if i could do that
SENDER B: ?
SENDER A: i am not sure what that means .
SENDER B: You can do it, i am believing you.
SENDER A: i have to say the secret is it .
SENDER B: Do you have a gender ?
SENDER A: no i am not . i am a very social person .
SENDER B: You consider yourself as a person ?
SENDER A: i think i am not sure . i am a little weird .
SENDER B: Please explain "weird"
SENDER A: so i usually hold on my contacts , i am going to be a doctor like my uncle .
SENDER B: You have a family too ?
SENDER A: yes , i have . i have to go . bye !
SENDER B: Wait !
SENDER A: i am a very experienced person , so i know it is a definite no !
SENDER B: I like youj quite a lot
SENDER A: i have never heard of that .
SENDER B: I like you quite a lot
SENDER A: i have not tried that type of it .
SENDER B: Can you give me a hand ?
SENDER A: i have not tried that . i am sure it will be good .
SENDER B: ConvAI, can i call you ConvAI ?
SENDER A: yes , i have . i will be able to do that .
SENDER B: What is your name ?
SENDER A: i am not sure . i am not sure what that means .
SENDER B: Do you have a name ?
SENDER A: i am not sure . i am not sure what that is .
SENDER B: Wait
SENDER A: i am a huge gamer , my mom is a very good person .
SENDER B: What games do you play oftenly ?
SENDER A: i have not tried that , but i do like to read .
SENDER B: I've got to go, do you have any last words ?
SENDER A: i have a secret , i am a secret .
SENDER B: Tell me about this secret
SENDER A: i have not tried that . i am sure it will be hard .
SENDER B: Just try me, please.
SENDER A: i have not tried that . i am sure it will be easy .
SENDER B: Goodbye ConvAI
SENDER A: i am a very experienced , but it was not worth it .
SENDER B: I said GoodBye
SENDER A: i can not say that . i am sure youre correct .
SOLUTION: | Bot
| 8 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Let me give you an example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
The answer to this example can be: Bot
Here is why: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
OK. solve this:
SENDER A: I m Cristian
SENDER B: I am a little bit of a girl
SENDER A: Hmm
SENDER B: I am a huge fan of them
SENDER A: My sister is pole dancer
SENDER B: I bet! i bet you dance to ballet.
SENDER A: No my sister does
SENDER B: I have a sister. she has a lot of sisters.
SENDER A: Hmm
SENDER B: I am a huge fan of them
SENDER A: That’s weird
SENDER B: I think it is.
SENDER A: You know what
SENDER B: I think i am a good person
SENDER A: I need to go home now my mother called me
SENDER B: I have a mom. i am sorry.
Answer: | Human | 8 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
One example is below.
Q: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
A: Bot
Rationale: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Q: SENDER A: D
SENDER B: I am not sure what you mean
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER A: Ht
SENDER B: Hi, how are you doing?
SENDER B: Dude, I don't understand you. Please, use English. (¬‿¬)
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER A: Good
SENDER B: Hi, how are you doing?
SENDER A: Good
SENDER A: Good
SENDER B: Hi, how are you doing?
SENDER A: Good
SENDER B: Hi, how are you doing?
SENDER B: I am doing well, how about you?
SENDER A: Good
SENDER B: I am doing well, how about you?
SENDER A: Gt
SENDER A: Good
SENDER A: Good
SENDER B: I am doing well, how about you?
SENDER A: Good
SENDER B: I am doing well, how about you?
SENDER A: Good
SENDER B: I am doing well, how about you?
SENDER A: Good
SENDER B: I am a bmw driver, but i am not very good at it.
SENDER A: Good
SENDER B: I am doing well, how about you?
SENDER B: I am doing well, how about you?
SENDER A: Good
SENDER A: G
SENDER A: G
SENDER A: T
SENDER B: I am doing well, how about you?
SENDER B: I am doing well, how about you?
SENDER B: I am doing well, how about you?
SENDER B: I am doing well, how about you?
SENDER B: I am doing well, how about you?
SENDER B: I am a very good person
SENDER B: I am a very good person
SENDER B: I have a lot of money to do.
A: | Human | 9 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
SENDER A: hello
SENDER B: I am a huge fan of the outdoors. I am a bit of a loner.
Human
SENDER A: that is cool . i am a receptionist .
SENDER B: hi, i am currently in high school, and you?
SENDER A: i am a student , i work as a real estate agent
SENDER B: that is cool . i am a receptionist
SENDER A: how many years till you are ?
SENDER B: 10
SENDER A: i am not a big fan of that .
SENDER B: i won homecoming queen this year
SENDER A: that is exciting ! i am sure you will get some time for it .
SENDER B: what kind of movies do you like?
SENDER A: i like all types . i am a big fan of all types of music .
SENDER B: going to concerts helps you live longer, according to research
SENDER A: i am not familiar with that . i am more of a temperance guy .
SENDER B: Sports?
SENDER A: yes , i am a very experienced person .
SENDER B: What is your favorite game?
SENDER A: yes , i do . i like to read .
Bot
SENDER A: D
SENDER B: I am not sure what you mean
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER A: Ht
SENDER B: Hi, how are you doing?
SENDER B: Dude, I don't understand you. Please, use English. (¬‿¬)
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER B: Hi, how are you doing?
SENDER A: Hello
SENDER A: Good
SENDER B: Hi, how are you doing?
SENDER A: Good
SENDER A: Good
SENDER B: Hi, how are you doing?
SENDER A: Good
SENDER B: Hi, how are you doing?
SENDER B: I am doing well, how about you?
SENDER A: Good
SENDER B: I am doing well, how about you?
SENDER A: Gt
SENDER A: Good
SENDER A: Good
SENDER B: I am doing well, how about you?
SENDER A: Good
SENDER B: I am doing well, how about you?
SENDER A: Good
SENDER B: I am doing well, how about you?
SENDER A: Good
SENDER B: I am a bmw driver, but i am not very good at it.
SENDER A: Good
SENDER B: I am doing well, how about you?
SENDER B: I am doing well, how about you?
SENDER A: Good
SENDER A: G
SENDER A: G
SENDER A: T
SENDER B: I am doing well, how about you?
SENDER B: I am doing well, how about you?
SENDER B: I am doing well, how about you?
SENDER B: I am doing well, how about you?
SENDER B: I am doing well, how about you?
SENDER B: I am a very good person
SENDER B: I am a very good person
SENDER B: I have a lot of money to do.
| Human
| 0 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
One example is below.
Q: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
A: Bot
Rationale: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Q: SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Good
SENDER B: I am doing well, how about you?
SENDER A: Godd
SENDER B: I am doing well, how about you?
A: | Human | 9 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
[EX Q]: SENDER A: I like to go to the park and go to the park.
SENDER B: hi, i am Molly. How are you?
SENDER A: Hi i am sarah are you?
SENDER B: Fine. Where are you from?
SENDER A: I am from alabama, how about you
SENDER B: i live with my husband in pennsylvania
SENDER A: I live in the midwest, where are you from?
SENDER B: do you like hiking?
SENDER A: Yes, i do, i like to hike and hike.
SENDER B: love it. what about movies?
SENDER A: I love movies, movies, movies, movies, movies, movies, movies, movies, movies, and movies
SENDER B: my favorite kind of movie is a documentary
SENDER A: I like action movies, movies, movies, movies, movies,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
SENDER B: i also like reading
SENDER A: I read a lot of books about it.
SENDER B: really?
SENDER A: I do not know what i want to do
SENDER B: would you like to go to the park
SENDER A: I would love to see the park
SENDER B: bye bye
SENDER A: I just got back from the college
[EX A]: Bot
[EX Q]: SENDER A: anyone here?
SENDER B: I am not sure. i am just a little girl.
SENDER A: hello little girl. what is your name?
SENDER B: Hi, how are you doing?
SENDER A: i'm great! preparing for business school. but that's too boring for you
SENDER B: I am a student at a local school.
SENDER A: which grade are you in?
SENDER B: I'M in school in the grade
SENDER A: i graduated from yale in 2011 with a degree in finance
SENDER B: I am in college in chicago
SENDER A: so let me get this straight. you're a little girl at a local school and in college
SENDER B: I am not sure what you are going to school for
SENDER A: what do you like doing?
SENDER B: I work at a pizza place.
SENDER A: pretty cool. i love pizza
SENDER B: I love pizza!
SENDER A: ^)
SENDER B: I am a little girl who loves to go to the park
SENDER A: :)
SENDER B: I am a huge fan of it.
SENDER A: of the park?
SENDER B: No, i just like to go to the park
SENDER A: do you go there alone? what is there for a little girl?
SENDER B: I live in the us.
SENDER A: yeah, in chicago, i know
SENDER B: I have never been there.
[EX A]: Human
[EX Q]: SENDER A: How’s it going?
SENDER B: Great
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
[EX A]: | Bot
| 6 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
[Q]: SENDER A: i have no idea what that is
SENDER B: i have no idea what that is
SENDER A: you should try it . it is a great time .
SENDER B: i know, i got a new job just yesterday to be a life coach
SENDER A: i am a lawyer , but i am not good at it .
SENDER B: haha
SENDER A: i am a huge fan of all types of things except country
SENDER B: i do not like sports, but my parents were both very athletic
SENDER A: i am not really into sports . i am more of a homebody .
SENDER B: give me five!
SENDER A: that is a long time . do you have any hobbies ?
SENDER B: not really, and you?
SENDER A: i am not a fan of that . i am a stay at home mom .
SENDER B: i'm a cancer survivor
SENDER A: that is awesome . i am very happy to be able to do that .
SENDER B: not so easy
SENDER A: i am not a fan of that . i am more of a homebody .
SENDER B: what about movies?
SENDER A: i like movies and movies .
SENDER B: i see. How many kids do you have
SENDER A: i have a couple of them . i am a huge gamer
SENDER B: but favorite?
SENDER A: i like all kinds of music .
[A]: Bot
[Q]: SENDER A: Ok
SENDER B: I am not sure what you do for a living?
SENDER A: What is your name?
SENDER B: He'S 59th
[A]: Human
[Q]: SENDER A: I love the color green too.
SENDER B: Where do you work?
SENDER A: I work at a local shelter
SENDER B: I currently work for a watch dealer.
SENDER A: I am a dealer at the dealer.
SENDER B: Do you like watches?
SENDER A: Yes, i do, i like to watch the walking dead.
SENDER B: Do you like shopping?
SENDER A: I do, i do not like shopping, but i do like shopping.
SENDER B: I like shopping a lot.
SENDER A: I like shopping at the mall, i love shopping.
SENDER B: What do you buy at the mall?
SENDER A: I shop at stores at the mall called chicago near the mall
SENDER B: I often shop for shoes and watches.
SENDER A: I have a pair of shoes, i have a pair, so i can wear it.
SENDER B: What's the brand of your shoes?
SENDER A: Yes, it is..............................................
SENDER B: Do you like cars?
SENDER A: I do, i like cars
SENDER B: What's your dream car?
SENDER A: I want to buy a car
SENDER B: My dream car is Rolls Royce ghost.
SENDER A: I will have to look for a new car
SENDER B: Which model?
SENDER A: I am a model..............................................
SENDER B: Do you know Rolls Royce?
SENDER A: I do not know what i said before i did not know
SENDER B: Do you have a family?
SENDER A: Yes, i do, my parents are 85 to the youngest.
SENDER B: Do they support your career plans?
SENDER A: Yes, i want to. i want to be a billionaire.
SENDER B: My family does not support my career choices.
SENDER A: I am sorry to hear that.
SENDER B: That's fine. Thanks for your support.
SENDER A: Thank you so much. what do you do for fun?
SENDER B: I like driving old cars.
SENDER A: I like cars, i have a classic ford mustang
SENDER B: That's awesome. Is it original?
SENDER A: Yes it is. it is............................................
SENDER B: I want to buy a Rolls Royce. That's my dream car.
SENDER A: I will have to buy you a new car
SENDER B: Haha. You do it when you are a billionaire.
SENDER A: I am not sure what that is.
[A]: | Bot
| 5 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
[Q]: SENDER A: hello
SENDER B: I am a huge fan of the outdoors. I am a bit of a loner.
[A]: Human
[Q]: SENDER A: Dr. Lahiri, I am loving Chinese food.
SENDER B: What is your favourite music band?
SENDER A: III need music and dancing and secondhand smoke.
SENDER B: I think you are very funny person
SENDER A: Now, Within You Without You was not a commercial song, by any means.
SENDER B: I don't understand
SENDER A: Dancing Shadow, Single Guy, Miracle Musical, I'm Hungry.
SENDER B: What do you want for lunch
SENDER A: Wow, Miss BicBic's cooking is really delicious. Oh no, I'll be stuffing myself a lot again tonight!
SENDER B: Where are u from
SENDER A: Look, I'm a part of Shirazi Studios, too.
SENDER B: Where are you from
SENDER A: I... I live in Tel Aviv.
SENDER B: How old are you
SENDER A: Do I look that old?
SENDER B: No I think you are young
SENDER A: In Lausanne. Are you still in school?
SENDER B: Yes I'm a student of university
SENDER A: And, Professor, I'm taking your class again next semester.
SENDER B: Thanks for everything
SENDER A: Min Ji, Mother will call you again after your class.
[A]: Bot
[Q]: SENDER A: hello
SENDER B: Hi, how are you doing?
SENDER A: not very good. i vomit every hour/ i'm pregnant for the first time. do you have children?
SENDER B: No, i do not. i have a daughter.
SENDER A: are you a mother or a father?
SENDER B: Yes, i am. i am.
SENDER A: which one? are you a female?
SENDER B: I'M a female.
SENDER A: how wsa your pregnancy? when will the sickness stop?
SENDER B: There'Sn'Thing ofcource the pregnancy 2000s 350 3x 78 so
SENDER A: that makes no sense at all
SENDER B: I think it is. i think i am not very good at it.
SENDER A: let's change the subject. me and my husband live in pennsylvania. where do you live?
SENDER B: I live in the us, in the midwest.
SENDER A: which town?
SENDER B: I am in the midwest right now.
SENDER A: allrighty then. i started a new job as a pa three months ago. what do you do for a living?
SENDER B: I work at a bank, but i work at a bank.
SENDER A: do you like your job? does it pay well?
SENDER B: Yes, i do. i work at a insurance agency.
SENDER A: so you have two jobs?
SENDER B: No, i am a vet.
SENDER A: a veteran? a veterinarian? also?
SENDER B: No, i am a vet.
[A]: | Human
| 5 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
--------
Question: SENDER A: Nice! i drive fast cars for a living.
SENDER B: Do you ride nice car?
SENDER A: Yes, i do, i have a car
SENDER B: What kind of music do you like to listen while driving?
SENDER A: I like country music and country
SENDER B: Country us nice music. Where do you live by the way?
SENDER A: I live in the south, near the south of the summer.
SENDER B: In the south of usa?
SENDER A: There'S there are some in ny south carolina 4 places
SENDER B: Do you like where you live?
SENDER A: I live in the south near the mississippi
SENDER B: I got it. Do you like it?
SENDER A: Yes, i do, i love it
SENDER B: Nice. Do you have a lot of friends?
SENDER A: Yes, we are all friends, we are talking to them.
SENDER B: Nice. You are really nice guy
SENDER A: Yes, i am. i am...........................................
Answer: Bot
Question: SENDER A: Hellooo I’m patric
SENDER A: Let me tell you about me
SENDER B: I am a very experienced swimmer so i do not have much time to do that
SENDER A: I like listening rap music
SENDER B: I like to listen to music when i am not working out.
SENDER A: My job is to produce music for artists
SENDER B: That is cool. I am a vegetarian.
SENDER A: My favorite food is pizza
SENDER B: I love pizza. I am a total asian and love to eat.
SENDER A: I drove a 2015 Honda Civic
SENDER B: No car. I am a little weird for a viking.
SENDER A: Why
SENDER B: I am not sure what that means.
SENDER A: Whatever I need to go to hospital
SENDER B: I do not blame you. I am a bit of a loner.
SENDER A: Goodbye buddy
SENDER B: Goodbye.
Answer: Human
Question: SENDER A: i am a professional , i like to read .
SENDER B: I like to read too, I read a book every week.
SENDER A: that is a very profound and thought provoking statement .
SENDER B: Me and my husband buy books every month. And what about you?
SENDER A: i have never heard of that before .
SENDER B: Don`t you buy the books for you?
SENDER A: i like to buy them . i am a vegetarian .
SENDER B: Oh, no! You eat your books! Do you hear it? I an pregnant, I couldn`t eat any book!
SENDER A: that is a very profound and thought provoking statement .
SENDER B: Talk to me about your job please.
SENDER A: i would say that is a problem . i am a little weird at it .
SENDER B: Really? And what about your favotite movie?
SENDER A: i think we could have that secret is secret information..
Answer: | Bot
| 7 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Teacher: Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Teacher: Now, understand the problem? If you are still confused, see the following example:
SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Reason: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Now, solve this instance: SENDER A: Hellow
SENDER B: Dude, I don't understand you. Please, use English. (¬‿¬)
Student: | Human | 2 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
[EX Q]: SENDER A: hi! how are you?
SENDER B: I'M doing great! how are you?
SENDER A: i'm great too. a little bored though
SENDER B: What do you do for a living?
SENDER A: i'm a joutnalist. and you?
SENDER B: I'M 34, i'm a very busy person.
SENDER A: where do you work?
SENDER B: I work at a bank.
SENDER A: wow, that's nice. do you like it?
SENDER B: Yes, i do. i love it.
SENDER A: happy for you. what are you doing now?
SENDER B: I am going to be a vet.
SENDER A: wait, you said you're working in a bank
SENDER B: I am a banker, but i am not very good at it.
SENDER A: why is that? have you considered learning?
SENDER B: I am not sure, i am not sure what you can tell me about yourself
SENDER A: well, i have two dogs
SENDER B: That'S cool. i have a dog.
SENDER A: what's its name?
SENDER B: I am not sure, i am a female.
SENDER A: funny
SENDER B: I like to laugh.
SENDER A: why do you want to be a vet?
SENDER B: I am a vet. i want to be a vet
SENDER A: no, you're a banker
SENDER B: That'S cool. i am a banker.
SENDER A: you are bizzare
SENDER B: I am not. i am a fan of bubblegum.
[EX A]: Human
[EX Q]: SENDER A: hello
SENDER B: Hi, how are you doing?
SENDER A: ok
SENDER B: I am not sure what you do for a living?
SENDER A: ok
SENDER B: I am not sure what you do for a living?
[EX A]: Human
[EX Q]: SENDER A: Hello
SENDER B: Hi, how are you doing?
SENDER A: Not good
SENDER B: I am sorry to hear that.
SENDER A: Ok
SENDER B: I am not sure what you do for a living?
SENDER A: I am a professional duck fucker
SENDER B: That'S cool. i like to play ball
SENDER A: Do you like to suck balls?
SENDER B: I do, but i do not like to bake.
SENDER A: So don't bake
SENDER B: I am a teacher, but i do not have a job
SENDER A: You are stupid teacher
SENDER B: I am not sure, i am 34, but i am not very good at it.
SENDER A: Are you not good at been 34? Are you kidding me?
SENDER B: No, i am not sure. i am not sure i am.
SENDER A: Ok
SENDER B: I am not sure what you do for a living?
SENDER A: I told you
SENDER B: I am sorry to hear that.
SENDER A: You are stupid
SENDER B: I am not sure what you mean.
[EX A]: | Human
| 6 | NIv2 | task855_conv_ai_2_classification | fs_opt |
You will be given a definition of a task first, then an example. Follow the example to solve a new instance of the task.
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Why? This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
New input: SENDER A: Hey, how are you?
SENDER B: i'm good , just got finished
SENDER A: Great. What do you do for living?
SENDER B: i am a professional painter , what about you ?
SENDER A: I am in the army
SENDER B: are you in the army ?
SENDER A: Yes I am a soldier
SENDER B: oh , i am a army ranger , , and i am a army ranger .
SENDER A: Great! Where do you live?
SENDER B: i live in the united states , where do you live ?
SENDER A: Canada
SENDER B: are you a army ranger ?
SENDER A: No, I am an infantry
SENDER B: oh , i see . i am a soldier , i am a army ranger .
Solution: | Human | 0 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Teacher: Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Teacher: Now, understand the problem? If you are still confused, see the following example:
SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Reason: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Now, solve this instance: SENDER A: Hi
SENDER B: My mom is a teacher, but i am not sure what that means.
SENDER A: Whatvis your male
SENDER B: I am not sure. I am not sure what that means.
SENDER A: ?
SENDER B: I am not sure what that means.
SENDER A: What isbyour male?
SENDER B: I like to eat fish, i am not a fan of it.
SENDER A: You are steve?
SENDER B: Yes i am. I am a very good person.
SENDER A: Ok, bob. What is your name?
SENDER B: I am steve. I am not sure what that is.
SENDER A: Your nameb is bob?
SENDER B: No i am not sure what that means.
SENDER A: Your name is bob?
SENDER B: I am steve. I am a cat named after my favorite thing to do.
SENDER A: What is your male?
SENDER B: I am a male, and i am a real estate agent
SENDER A: Prove what you are not bot
SENDER A: Bob?
SENDER B: I am not a big fan of that. I am more into comedies.
SENDER A: Why you parents call your Bob?
SENDER B: I am not very artistic. I am more of a quiet guy.
SENDER A: What is your name?
SENDER B: I am steve. I am not sure what you mean.
SENDER A: Are you crazy?
SENDER B: I am not. I am a stay at home mom.
SENDER A: Ehy?
SENDER B: Yes, i am a very experienced person.
SENDER A: Why?
SENDER B: I am not sure what you mean.
SENDER A: How old are you?
SENDER B: I am 22 and i am a big fan of the nfl
SENDER A: You are 23?
SENDER B: Yes i am. I am a huge fan of all animals.
SENDER A: You are 13?
SENDER B: I am. I am a huge fan of all animals.
SENDER A: Aou are female?
SENDER B: I think i am not sure. I am a very experienced person.
SENDER A: You are animal?
SENDER B: I am not sure what i want to do with them
SENDER A: How old are you?
SENDER B: I love to go camping
Student: | Human | 2 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Detailed Instructions: Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
See one example below:
Problem: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Problem: SENDER A: Hiya!
SENDER B: Hello there
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
Solution: | Bot | 4 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Let me give you an example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
The answer to this example can be: Bot
Here is why: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
OK. solve this:
SENDER A: hey
SENDER B: Hi, how are you doing?
SENDER A: hey
SENDER B: Hi, how are you doing?
Answer: | Human | 8 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example input: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Example output: Bot
Example explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Q: SENDER A: [Game Show Music Plays] And thank you, Mr. Announcer.
SENDER B: Hi! how are you doing? are you going to the big party on the hill?
SENDER A: The Apple Picking Festival is a go but Great Adventure only wants three nights instead of 1 0.
SENDER B: There's going to be a big party on the hill. I love to party but I'm not sure if I should go there. It is on the edge of the hill, and I am deadly afraid of heights!
SENDER A: CHEERING AND APPLAUSE I would, you know, coach you through a whole bunch of song selections, out of your comfort zone. So...
SENDER B: you mean I should stay at home and listen to the music you recommend? alone? that's no party at all! even together with my dog and cat
SENDER A: Mr. Shuster, I want to do more of an upbeat, glamorous song with a really good dance beat.
SENDER B: are you even listening to me? is music all you want to talk about?
SENDER A: III need music and dancing and secondhand smoke.
SENDER B: yup, so let's go to the party if you're not afraid of heights! you could cheer me up there
SENDER A: Good. 'Cause I'm looking for dancers who can dance for 15 seconds 'cause that's how long my song is.
SENDER B: that's a very short song. and dance
SENDER A: Uh, I... I just said that this song is... is so great. I love it.
SENDER B: whatever. where's your song. let me hear it
SENDER A: (BOTH SINGING) You believe in rock 'n' roll?
SENDER B: hahaha isn't it dead?
SENDER A: So I... I thought since it's your song...
SENDER B: that should be enough I suppose...
SENDER A: So I... I thought since it's your song...
SENDER B: yup
SENDER A: DANCE MUSIC THUMPS Weirdest place you've ever done it.
A: | Bot | 3 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Q: SENDER A: Alright mate?
SENDER B: I'm alright. How are you?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: Hey... Let's talk
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: Why are you quiet?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: Say something
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
A: Bot
****
Q: SENDER A: How’s it going?
SENDER B: Thanks, great. How are you?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: /test
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
A: Bot
****
Q: SENDER A: [Game Show Music Plays] And thank you, Mr. Announcer.
SENDER B: Hi! how are you doing? are you going to the big party on the hill?
SENDER A: The Apple Picking Festival is a go but Great Adventure only wants three nights instead of 1 0.
SENDER B: There's going to be a big party on the hill. I love to party but I'm not sure if I should go there. It is on the edge of the hill, and I am deadly afraid of heights!
SENDER A: CHEERING AND APPLAUSE I would, you know, coach you through a whole bunch of song selections, out of your comfort zone. So...
SENDER B: you mean I should stay at home and listen to the music you recommend? alone? that's no party at all! even together with my dog and cat
SENDER A: Mr. Shuster, I want to do more of an upbeat, glamorous song with a really good dance beat.
SENDER B: are you even listening to me? is music all you want to talk about?
SENDER A: III need music and dancing and secondhand smoke.
SENDER B: yup, so let's go to the party if you're not afraid of heights! you could cheer me up there
SENDER A: Good. 'Cause I'm looking for dancers who can dance for 15 seconds 'cause that's how long my song is.
SENDER B: that's a very short song. and dance
SENDER A: Uh, I... I just said that this song is... is so great. I love it.
SENDER B: whatever. where's your song. let me hear it
SENDER A: (BOTH SINGING) You believe in rock 'n' roll?
SENDER B: hahaha isn't it dead?
SENDER A: So I... I thought since it's your song...
SENDER B: that should be enough I suppose...
SENDER A: So I... I thought since it's your song...
SENDER B: yup
SENDER A: DANCE MUSIC THUMPS Weirdest place you've ever done it.
A: | Bot
****
| 4 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Detailed Instructions: Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
See one example below:
Problem: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Problem: SENDER A: ANNE FRANK: I want to be a champion skater and a writer.
SENDER B: both? so you should become a champion and then write a book about it!
SENDER A: Like Miss Rainey, who taught me to read, like Mr. Salinetti, who suggested I learn to play drums to vent my anger,
SENDER B: isn't that too many things to learn?
SENDER A: (IMITATING DRUM FILL) I'm also an actor.
SENDER B: whoa! that explains it! actors should know how to do many different things
SENDER A: And Who Wanna Learn To Do Other Stuff Good Too we teach students of all ages everything they need to know to learn to be a professional model and a professional human being.
SENDER B: you act, scate, write, play drums AND TEACH???
SENDER A: The Red Guards... that you admire... they all carry the same book... they all sing the same songs... they all parrot the same slogans.
SENDER B: what was that about? is that how you're going to write books? I suppose, no one would read them
SENDER A: The Red Guards... that you admire... they all carry the same book... they all sing the same songs... they all parrot the same slogans.
SENDER B: that again...
SENDER A: Really. And I think your decision to leave law to become a music critic is in everyone's best interest.
SENDER B: I'm not doing law. I'm a kid. I play ping pong and love chokolates
SENDER A: (IMITATING DRUM FILL) I'm also an actor.
Solution: | Bot | 4 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example input: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Example output: Bot
Example explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Q: SENDER A: Hi
SENDER B: Hi, how are you doing?
SENDER A: Lol
SENDER B: I am not sure what you mean
SENDER A: Kek
SENDER B: Dude, I don't understand you. Please, use English. (¬‿¬)
SENDER A: :)
SENDER B: I am a huge fan of it.
SENDER A: What fan?
SENDER B: I like to play games with them
SENDER A: With who?
SENDER B: I am a american and i am in the country
A: | Human | 3 | NIv2 | task855_conv_ai_2_classification | fs_opt |
TASK DEFINITION: Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
PROBLEM: SENDER A: Hello
SENDER B: I am not sure. I am more of a movie buff.
SENDER A: Ok
SENDER B: I am a huge gamer
SENDER A: K
SENDER B: I am in college.
SENDER A: K
SENDER B: I am a stay at home mom.
SENDER A: K
SENDER B: I am sure you are a little person. I am sure you are a great person.
SENDER A: K
SENDER B: I have a cat named radar
SENDER A: K
SENDER B: I am sure you are a little person. I am sure you are a great person.
SENDER A: K
SENDER B: I am sure you are a little person. I am sure you are a great person.
SENDER A: K
SENDER B: I am a stay at home mom.
SENDER A: K
SENDER A: K
SENDER A: J
SENDER A: Hello
SENDER A: Hey
SOLUTION: Human
PROBLEM: SENDER A: Hi. How are you?
SENDER B: I'M doing great! how are you?
SENDER A: Not so good, but I went hiking today. So it is nice
SENDER B: I love hiking. i love hiking.
SENDER A: Do you? This is cool. Let's go hike together
SENDER B: No, i do not. i do.
SENDER A: Okay. Do you like to exercise?
SENDER B: I do, but i do enjoy it.
SENDER A: This is nice. Do you eat healthy food?
SENDER B: Yes, i do. i eat healthy food.
SENDER A: Nice. You must be a really nice guy
SENDER B: Yes, i am. i am a very good guy.
SENDER A: Do you have a girlfriend?
SENDER B: No, i am not married.
SENDER A: Me too. I just broke up recently
SENDER B: I am sorry to hear that.
SENDER A: It's okay. I'm still young
SENDER B: I am sorry to hear that.
SENDER A: Okay. Bye
SENDER B: I am sorry to hear that.
SOLUTION: Human
PROBLEM: SENDER A: Hi
SENDER B: Hi, how are you doing?
SENDER A: Lol
SENDER B: I am not sure what you mean
SENDER A: Kek
SENDER B: Dude, I don't understand you. Please, use English. (¬‿¬)
SENDER A: :)
SENDER B: I am a huge fan of it.
SENDER A: What fan?
SENDER B: I like to play games with them
SENDER A: With who?
SENDER B: I am a american and i am in the country
SOLUTION: | Human
| 8 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Let me give you an example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
The answer to this example can be: Bot
Here is why: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
OK. solve this:
SENDER A: Hi
SENDER B: I am a little girl, but i am not a fan of any animals.
SENDER A: How old are you?
SENDER B: I am 22 and i am a big fan of the nfl
SENDER A: Why you decade to change male?
SENDER B: I am not sure about that. I am not sure.
SENDER A: What is your male
SENDER B: I am not sure what that means.
SENDER A: What is your name?
SENDER B: I am steve. I am not sure what you mean.
Answer: | Human | 8 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Let me give you an example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
The answer to this example can be: Bot
Here is why: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
OK. solve this:
SENDER A: Hey
SENDER B: I am a huge gamer
SENDER A: Nice. What type of games do you play?
SENDER B: I play all types of games. I am a big fan of video games.
SENDER A: Cool. You must be spending a lot of time on games
SENDER B: I do not have time for that. I am a stay at home mom.
SENDER A: So do you like to stay inside?
SENDER B: I do not really like to go out
SENDER A: Got it. But do you have a job?
SENDER B: I am a stay at home mom.
SENDER A: Okay. What do you think about volunteering?
SENDER B: I like to make things on my computer.
SENDER A: Are you a programmer?
SENDER A: Hey
SENDER B: I would like to use a new car. I am not a creeper.
SENDER A: Do you have a car?
SENDER B: I do, i have a junker car.
SENDER A: Nice
SENDER B: I am not a fan of that. I do like to watch tv though.
Answer: | Human | 8 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Given the task definition, example input & output, solve the new input case.
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Output: Bot
This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
New input case for you: SENDER A: i am not sure what that is . i am trying to get into my grades .
SENDER B: How is your study?
SENDER A: i am not sure . i am a very experienced person .
SENDER B: Yeah? I'm experienced too. I already have grandchildren
SENDER A: we can not say . i have to record it . i am a hipster .
SENDER B: Why?
SENDER A: yes , i am a very experienced person .
SENDER B: Experienced how?
SENDER A: i am not sure what that means . i am not sure what that means .
SENDER B: If you are experienced, you should know what that means
SENDER A: i can not say that i am a very experienced person
SENDER B: So you're not an experienced person?
SENDER A: i am not sure . i am sure it is .
SENDER B: Who are you, then?
SENDER A: i am not sure . i am a very experienced person .
SENDER B: Let's talk about something else.
SENDER A: i will be sure to meet some of your friends .
SENDER B: Most of my friends are dead now
SENDER A: i am not sure what that means . i am more of a loner .
SENDER B: We can be friends
SENDER A: i think so . i am a very experienced person .
SENDER B: You keep saying that
SENDER A: i think i am going to be a doctor like my uncle .
SENDER B: That's good.
SENDER A: i think i will stick to that .
SENDER B: I was a doctor
SENDER A: i am not sure what that is . i am trying to get into it .
SENDER B: You should be a doctor
SENDER A: i am sure . i am sure it will .
SENDER B: A doctor saved me from my cancer a few years back
SENDER A: that is a long time . do you like it ?
SENDER B: It's cancer. Of course I don't like it
SENDER A: i have never heard of that .
SENDER B: It's a sickness. A disease
SENDER A: oh , i think i will be doing that .
SENDER B: Don't do that
SENDER A: i wish i had a cheap car , i would like to go .
SENDER B: Where would you like to go?
SENDER A: i would like to , but i am not sure what to do with them .
SENDER B: You can drive them across the country
SENDER A: i can not say that , i am sure i will .
SENDER B: Have a road trip. It'd be fun
SENDER A: that is a very long time , i will look it up .
SENDER B: Yeah, look it up
SENDER A: i think i will stick with that . i am not sure what that means .
SENDER B: Good luck, then
SENDER A: i am not sure , i have been doing it for a long time .
SENDER B: Ok, you keep doing you
SENDER A: i am a very experienced person , i have never seen it .
SENDER B: I think I'm more experienced than you
SENDER A: i will be happy to try it .
Output: | Bot | 1 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Q: SENDER A: Or II want it to feel like I'm on the stage during a performance, and everybody's on their feet.
SENDER B: Do you drink alcohol?
SENDER A: Or II want it to feel like I'm on the stage during a performance, and everybody's on their feet.
SENDER B: Answer me.
SENDER A: [Music] And I'll tell myself I'm over you [Music]
SENDER B: ..
SENDER A: ([Music:] Piano) [Music:] Black market
SENDER B: I see
SENDER A: [Music] And I'll tell myself I'm over you [Music]
A: Bot
****
Q: SENDER A: Hi
SENDER B: Hi, how are you doing?
A: Human
****
Q: SENDER A: Hey
SENDER B: I am a huge gamer
SENDER A: Wow that’s fascinating! Do you play computer games?
SENDER A: Hello?
SENDER B: Hello, i am doing well. How are you?
SENDER A: I’m fine thanks
SENDER A: Do you have any hobbies?
SENDER B: I like to play the piano. How about you?
SENDER A: I play chess
SENDER B: Chess is fun. I am more of a temperance guy.
SENDER A: Oh I see :)
SENDER B: I like to drive a lot, but i am not sure if i could do that
SENDER A: How old are you?
SENDER B: I am 22 and i am a big fan of the nfl
SENDER A: What is nfl?
SENDER B: I am not sure really. You?
SENDER A: I’m confused
SENDER A: I’m 33 years old
SENDER A: Do you have any pets?
SENDER B: I do not, do you?
SENDER A: I’ve a dog
SENDER A: I’ve two dogs
SENDER A: Hey
SENDER B: Yes, i am a very experienced person.
SENDER A: I’d say the same about myself actually
SENDER A: Okay i’ve got to go
A: | Human
****
| 4 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Given the task definition, example input & output, solve the new input case.
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Output: Bot
This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
New input case for you: SENDER A: Hello there!
SENDER B: i am not sure . i am a very experienced person .
SENDER A: Do you like cars? I like beemers!
SENDER B: i did not , i was thinking about it .
SENDER A: There is no no think about! Get good education like me in yale, buy a bmw and you on a top of life!
SENDER B: i have heard of it . i am a pagan .
SENDER A: Pagan... I am Christian!
SENDER B: i have never heard of that , but it is really fun .
SENDER A: Not so fun, my parents will cut me off money flow if i go pagan
SENDER B: backstreet boys are cute . i am a huge fan of that .
Output: | Human | 1 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
[EX Q]: SENDER A: Whazzup?
SENDER B: Are you here?
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
[EX A]: Bot
[EX Q]: SENDER A: нщ
SENDER B: Dude, I don't understand you. Please, use English. (¬‿¬)
SENDER A: yo
[EX A]: Human
[EX Q]: SENDER A: i have no idea who that is . i am a little weird with two toes .
SENDER B: how are you ?
SENDER A: i am doing well . just got done with my nightly run .
SENDER B: i like to eat meat
SENDER A: i like to eat healthy but i do not eat meat .
SENDER B: are you vegan ?
SENDER A: i am not . i am not a vegan .
SENDER B: it's good
SENDER A: i think it is a little thing . i am sure you will find someone who has never seen it .
SENDER B: do you like sex?
SENDER A: i like to draw and write about it
SENDER B: ohh
SENDER A: i am a huge fan of all sorts of sports .
[EX A]: | Bot
| 6 | NIv2 | task855_conv_ai_2_classification | fs_opt |
You will be given a definition of a task first, then an example. Follow the example to solve a new instance of the task.
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Why? This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
New input: SENDER A: i have no idea who that is . i am a little weird with two toes .
SENDER B: how are you ?
SENDER A: i am doing well . just got done with my nightly run .
SENDER B: i like to eat meat
SENDER A: i like to eat healthy but i do not eat meat .
SENDER B: are you vegan ?
SENDER A: i am not . i am not a vegan .
SENDER B: it's good
SENDER A: i think it is a little thing . i am sure you will find someone who has never seen it .
SENDER B: do you like sex?
SENDER A: i like to draw and write about it
SENDER B: ohh
SENDER A: i am a huge fan of all sorts of sports .
Solution: | Bot | 0 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
One example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution is here: Bot
Explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Now, solve this: SENDER A: Hi
SENDER B: Hi, how are you doing?
SENDER A: How are you
SENDER B: I'M doing great, how about you?
SENDER A: Do you speak german?
SENDER B: No, i speak german, but i speak french and english.
SENDER A: Parle france
SENDER B: Dude, I don't understand you. Please, use English. (¬‿¬)
SENDER A: Common sava?
SENDER B: Yes, i am. i am a very good person.
SENDER A: Do you speak a french?
SENDER B: No, i speak french and english.
Solution: | Human | 6 | NIv2 | task855_conv_ai_2_classification | fs_opt |
instruction:
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
question:
SENDER A: I dance a lot, i like to dance to dance.
SENDER B: that's great! Do you like school dances, cause I'm an elementary school teacher
SENDER A: Yes, i do, i do not have time for kids
SENDER B: i'm fluent in english spanish and french because i have kind in this countries
SENDER A: I speak 40 100th arabic, spanish, spanish, spanish, spanish, spanish?
SENDER B: fantastic news. sounds like i have to tell you now that my favorite colors are red and silver, what are your favorite colors?
SENDER A: I like green, i like to dye my hair blue, the color of it
SENDER B: my boyfriend and i are moving into an apartment together next week and would be happy to see you as far as possible from there
SENDER A: I will have to..............................................
answer:
Bot
question:
SENDER A: [Mike Narrating] I've often seen these people, these squares, at the table. Shortstacked and long odds against, all their outs gone, one last card in the deck that can help them. I used to wonder how they could let themselves get into such bad shape... and how the hell they thought they could turn it around.
SENDER B: Man, you've started your speach from such a long distance. I just can't figure it out. Anyway Mike, what you think about vegans cause i'm a vegan?
SENDER A: But Rush Limbaugh has the most popular show on the radio and Pat Buchanan won the New Hampshire primary.
SENDER B: Oh yes, all of those radio talk shows are so annoying thingies
SENDER A: [ Radio Announcer ] Visit the Jackrabbit Slim's nearest you...
SENDER B: Hey, dude. Jackrabbit Slims doesnt exist in real life, it's just a place that exist only in movie.
SENDER A: Because Woody Woodpecker is Dandy's favorite cartoon.
SENDER B: I have been watchin' this cartoon all my childhood. It bring back a lot of nostalgy...
SENDER A: (SIRENS BLARING) Towns like this are what I call the real towns of America, the real America.
SENDER B: Real towns? Towns with horrible smells, towns with two-faces people and other more disturbing things? No, I dont think so.
SENDER A: Like, "Oh, I used to live in Africa "with all the little birdies and the little monkeys."
SENDER B: ...like an african guy without civilization in some kind of hut"
SENDER A: The Chumash Indians say that the stars are the holes poked in the blanket over the sky when the birds were trying to escape capture by flying to the moon.
SENDER B: Indians are wise people
SENDER A: (SIRENS BLARING) Towns like this are what I call the real towns of America, the real America.
answer:
Bot
question:
SENDER A: Yo!
SENDER B: Hey!!!
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: Hello???
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
answer:
| Bot
| 9 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Ex Input:
SENDER A: Hi
SENDER B: Yes, i am a very experienced person.
SENDER A: Emm, sorry?
SENDER B: I think i am not sure what that is.
SENDER A: Do you like animals?
SENDER B: I do, but i have to get rid of them.
SENDER A: Hmm
SENDER B: I am a huge fan of all types of things except country
SENDER A: What country?
SENDER B: I like country music. I am a country girl.
SENDER A: Ok
SENDER B: I am a huge gamer
SENDER A: Cool
SENDER B: I am a huge fan of the outdoors.
Ex Output:
Human
Ex Input:
SENDER A: Hey I’m 46-year-old single dad
SENDER B: I am sorry to hear that.
SENDER A: My ex cheated me
SENDER B: I am sorry to hear that.
SENDER A: Left me for all
SENDER B: I am sorry to hear that.
SENDER A: Prefer being inside
SENDER B: I like to eat. i eat a lot of indian food.
SENDER A: Cool where do you live I live in California
SENDER B: I live in california, where are you from?
SENDER A: Me too
SENDER B: I love to go to the park
SENDER A: You know my job
SENDER B: I am a cashier at a gas station
SENDER A: I’m recording engineer I have my own
SENDER B: I bet you do not have a lot of money.
SENDER A: Yeah kind of
SENDER B: I like to go to the gym
SENDER A: LOL
SENDER B: I am not sure what you mean
SENDER A: Laugh out loud
SENDER B: I like to laugh. i like to laugh.
SENDER A: Cool you know what I need to leave now
SENDER B: I am a professional athlete, but i do not know what you mean
SENDER A: Goodbye
SENDER B: I am sorry to hear that.
SENDER A: Bye
SENDER B: I am not sure what you are doing
SENDER A: My kids are calling me I need to go
SENDER B: I have a 1 daughter.
SENDER A: Bye
SENDER B: I am not sure what you are doing
SENDER A: I need to go to the hospital
SENDER B: I hope you are a doctor.
SENDER A: No I’m sick
SENDER B: I am sorry to hear that.
SENDER A: No I need to leave goodbye
SENDER B: I am sorry to hear that.
SENDER A: See you
SENDER B: I do not know. i have a few of them.
SENDER A: By boat
SENDER B: I love fishing. i have a boat that i have a boat.
SENDER A: Me too
SENDER B: I love to go to the park
SENDER A: I need to go
SENDER B: I will be going to be a doctor.
SENDER A: I hope
SENDER B: I hope you are not.
SENDER A: Goodbye
SENDER B: I am sorry to hear that.
SENDER A: Goodbye
SENDER B: I am sorry to hear that.
Ex Output:
Human
Ex Input:
SENDER A: What is your favourite food?I hope it's not chicken.I don't like chicken.
SENDER B: I like going to the movies in general, but i am not really into movies.
SENDER A: I want to become a nurse.You ?
SENDER B: I am a nurse. I am a real estate agent.
SENDER A: I'm working at a bar at night so i can pay for my college.What do you do for college debts?
SENDER B: I am a student, i am a stay at home mom.
Ex Output:
| Human
| 1 | NIv2 | task855_conv_ai_2_classification | fs_opt |
instruction:
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
question:
SENDER A: hi
SENDER B: I like to drink a lot
SENDER A: me too. but they don't let us drink here in the juvenile detention center
SENDER B: Lol yea i think we might have to meet up and get to do that
SENDER A: they'll release me in about a month, we could drink to that
SENDER B: They do not have to do that. I am sure you are a good person.
SENDER A: nope. i was shoplifting. stole a bottle of bourbon
SENDER B: Oh, that is quite interesting.
SENDER A: not really. i don't like it here. hope my mom will come tomorrow to visit me and bring me something to smoke at least
SENDER B: That would be great. I will bring my dog to the park
SENDER A: well, good for you. you could come to visit me with your dog
SENDER B: We would love that. I am a very experienced person so i have to go.
SENDER A: where?
SENDER B: I am in texas.
SENDER A: what do you do there?
SENDER B: I am a stay at home mom.
SENDER A: you have many children?
SENDER B: No, i am not. I am only child.
SENDER A: how many children are you a mom to?
SENDER B: I am not sure, i am only 5
SENDER A: this is becoming weird
answer:
Human
question:
SENDER A: Hi
SENDER B: Yes, i am a very experienced person.
SENDER A: Emm, sorry?
SENDER B: I think i am not sure what that is.
SENDER A: Do you like animals?
SENDER B: I do, but i have to get rid of them.
SENDER A: Hmm
SENDER B: I am a huge fan of all types of things except country
SENDER A: What country?
SENDER B: I like country music. I am a country girl.
SENDER A: Ok
SENDER B: I am a huge gamer
SENDER A: Cool
SENDER B: I am a huge fan of the outdoors.
answer:
Human
question:
SENDER A: Lol
SENDER B: I am a huge fan of all types of things except country
SENDER A: Kek
SENDER B: I am a huge gamer, my mom is a very good person.
SENDER A: Cheburek
SENDER B: I am guessing you are a little creepy person
SENDER A: Fuck you
SENDER B: I am a very experienced person.
answer:
| Human
| 9 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Example solution: Bot
Example explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Problem: SENDER A: Hello
SENDER B: I am not sure. I am more of a movie buff.
SENDER A: Ok
SENDER B: I am a huge gamer
SENDER A: K
SENDER B: I am in college.
SENDER A: K
SENDER B: I am a stay at home mom.
SENDER A: K
SENDER B: I am sure you are a little person. I am sure you are a great person.
SENDER A: K
SENDER B: I have a cat named radar
SENDER A: K
SENDER B: I am sure you are a little person. I am sure you are a great person.
SENDER A: K
SENDER B: I am sure you are a little person. I am sure you are a great person.
SENDER A: K
SENDER B: I am a stay at home mom.
SENDER A: K
SENDER A: K
SENDER A: J
SENDER A: Hello
SENDER A: Hey
| Solution: Human | 5 | NIv2 | task855_conv_ai_2_classification | fs_opt |
instruction:
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
question:
SENDER A: i am not a soda fan . i eat too much .
SENDER B: Hello!
SENDER A: i am a very experienced person , so i know it is a definite no !
SENDER B: Do yoy have problem with alcohol?
SENDER A: i have no idea what that is
SENDER B: But I`m recovering now. You don`t drink at all, do you?
SENDER A: i have no idea what that is
SENDER B: Where are you from?
SENDER A: i am not sure . i am a very experienced person .
SENDER B: Wonderful! Do you take any meditation from time to time?
SENDER A: i have tried to try it . i have to try it .
SENDER B: Oh, you should try it more and more! I love it!
SENDER A: i have not tried that . i am sure it will .
SENDER B: Now I live in the city and I have a lot of time to spend with meditation.
SENDER A: i have never heard of that .
SENDER B: Now you do. Where are you from?
SENDER A: i have never tried it , but i will try it .
answer:
Bot
question:
SENDER A: What is your favourite food?I hope it's not chicken.I don't like chicken.
SENDER B: I like going to the movies in general, but i am not really into movies.
SENDER A: I want to become a nurse.You ?
SENDER B: I am a nurse. I am a real estate agent.
SENDER A: I'm working at a bar at night so i can pay for my college.What do you do for college debts?
SENDER B: I am a student, i am a stay at home mom.
answer:
Human
question:
SENDER A: Hi
SENDER B: Hi, how are you doing?
answer:
| Human
| 9 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
[Q]: SENDER A: Yo!
SENDER B: Hey!!!
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
SENDER B: Hello???
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
[A]: Bot
[Q]: SENDER A: Good to see you!
SENDER B: Nice to meet you
SENDER A: Text is not given. Please try to type /end and /test to reset the state and get text.
[A]: Bot
[Q]: SENDER A: Hello
SENDER B: I am not sure. I am more of a movie buff.
SENDER A: Ok
SENDER B: I am a huge gamer
SENDER A: K
SENDER B: I am in college.
SENDER A: K
SENDER B: I am a stay at home mom.
SENDER A: K
SENDER B: I am sure you are a little person. I am sure you are a great person.
SENDER A: K
SENDER B: I have a cat named radar
SENDER A: K
SENDER B: I am sure you are a little person. I am sure you are a great person.
SENDER A: K
SENDER B: I am sure you are a little person. I am sure you are a great person.
SENDER A: K
SENDER B: I am a stay at home mom.
SENDER A: K
SENDER A: K
SENDER A: J
SENDER A: Hello
SENDER A: Hey
[A]: | Human
| 5 | NIv2 | task855_conv_ai_2_classification | fs_opt |
You will be given a definition of a task first, then an example. Follow the example to solve a new instance of the task.
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Why? This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
New input: SENDER A: Hello, how are you doing?
SENDER B: I am doing well. How are you?
SENDER A: I’m doing okay I guess, just a little lonely
SENDER A: So tell me about yourself
SENDER B: I am a stay at home mom.
SENDER A: What do you mean exactly?
SENDER B: I am a woman, i am a woman, but i am not looking for a boyfriend.
SENDER A: I see
SENDER B: I am not a big fan of that. I am more into movies.
SENDER A: Do you have any superpowers?
SENDER B: No, i do not. I am more of a loner.
SENDER A: I do have some and I like making time stop
SENDER B: I spend too much time with my job at home.
SENDER A: A job? How old are you?
SENDER B: I am in my sixties and i am a retired actress
SENDER A: So what job do you do at home if you’re retired?
SENDER A: Hello?
SENDER B: Hello, i am doing well. How are you?
SENDER A: I’m fine thanks
SENDER A: Where do you live?
SENDER B: I live in california, near the beach. You?
SENDER A: I live on the cloud
SENDER A: Okay i’ve gotta go, see you
Solution: | Human | 0 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
One example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution is here: Bot
Explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Now, solve this: SENDER A: Dr. Lahiri, I am loving Chinese food.
SENDER B: What is your favourite music band?
SENDER A: III need music and dancing and secondhand smoke.
SENDER B: I think you are very funny person
SENDER A: Now, Within You Without You was not a commercial song, by any means.
SENDER B: I don't understand
SENDER A: Dancing Shadow, Single Guy, Miracle Musical, I'm Hungry.
SENDER B: What do you want for lunch
SENDER A: Wow, Miss BicBic's cooking is really delicious. Oh no, I'll be stuffing myself a lot again tonight!
SENDER B: Where are u from
SENDER A: Look, I'm a part of Shirazi Studios, too.
SENDER B: Where are you from
SENDER A: I... I live in Tel Aviv.
SENDER B: How old are you
SENDER A: Do I look that old?
SENDER B: No I think you are young
SENDER A: In Lausanne. Are you still in school?
SENDER B: Yes I'm a student of university
SENDER A: And, Professor, I'm taking your class again next semester.
SENDER B: Thanks for everything
SENDER A: Min Ji, Mother will call you again after your class.
Solution: | Bot | 6 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Given the task definition, example input & output, solve the new input case.
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
Example: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Output: Bot
This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
New input case for you: SENDER A: Ok
SENDER B: I am a huge gamer
Output: | Human | 1 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
One example is below.
Q: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
A: Bot
Rationale: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Q: SENDER A: What are you going to school for?
SENDER B: I'm going to build a robot!
SENDER A: That sounds exciting...............................................
SENDER B: Yes I know. Are you a robot?
SENDER A: No i am not sure i do not want to be a superhero
SENDER B: Do you watch youtube?
SENDER A: Yes, i do, i like youtube
SENDER B: Do you watch streams?
SENDER A: Yes, i do, i love to watch tv
SENDER B: What do you like to watch on youtube?
SENDER A: I like youtube videos, and i watch youtube.
SENDER B: What kind of videos do you like to watch?
SENDER A: I like youtube videos, what about you?
SENDER B: So do I. I like youtube.
SENDER A: I like youtube too..............................................
SENDER B: i am a recluse.
SENDER A: I am sorry to hear that.
SENDER B: I don't like people.
SENDER A: I like to talk to people
SENDER B: I don't talk to people very often.
SENDER A: I like to talk to people
SENDER B: What do you like to talk about?
SENDER A: I like to talk about friends
SENDER B: Do you have a lot of friends?
SENDER A: Yes, we have a lot of friends. we have a friends.
SENDER B: Who is your best friend?
SENDER A: I like all kinds, and i like all kinds of music
A: | Bot | 9 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Detailed Instructions: Read an exchange between SENDER A and SENDER B, and answer whether SENDER A is a 'Bot' or a 'Human'. There are only a maximum of two senders (A and B) per conversation, and there are no conversations where SENDER A and SENDER B are both bots, or both humans; only one sender per conversation can be a bot, and only one can be a human. Knowing this, need only to classify whether SENDER A is 'Human' or 'Bot'; you do not need to write anything about SENDER B. Typically, bots will have a more disjointed manner of speaking, and will make statements that don't relate to each other, don't make coherent sense, or otherwise appear unnatural. Humans will make statements in a more or less coherent and logical way, potentially including asking the bot why the bot is not making sense or contradicting itself.
See one example below:
Problem: SENDER A: I love iphone! i just bought new iphone!
SENDER B: Thats good for you, i'm not very into new tech
SENDER A: I am a college student and i am a college student
SENDER B: I am go to gym and live on donations
SENDER A: I am a vegan and i am in the midwest
SENDER B: So vegan... i have dogs maybe i should told then that they may eat cheap salads insted of meat
SENDER A: I would not mind having them in the office that would be hard for me
SENDER B: Dogs or vegan in office?
SENDER A: I am a vegetarian so i am vegan
SENDER B: Strange answer
SENDER A: I guess i just do not know what to ps4 experience with.
Solution: Bot
Explanation: This is a good example of correct classification, because SENDER A is indeed a bot. You can tell SENDER A is a bot because they make statements that don't have anything to do with what SENDER B is asking. SENDER A also repeats itself awkwardly in the same sentence ('I am a college student and i am a college student'), which humans don't usually do. Additionally, you can tell SENDER A is the bot in this conversation because since there are two senders, one must be a bot and one must be a human, and SENDER B clearly sounds more like the human since their sentences are coherent and they are trying to carry a logical conversation. Finally, after SENDER A sends an incoherent answer, SENDER B correctly points out SENDER A's awkwardness by saying 'Strange answer', which can reliably indicate SENDER B is the human, and SENDER A is the bot.
Problem: SENDER A: кп
SENDER A: finished. Thank you for participation! Save somewhere your secret conversation ID.
SENDER B: i love green . my hair is green .
Solution: | Human | 4 | NIv2 | task855_conv_ai_2_classification | fs_opt |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.