@inproceedings{lee-etal-2022-personachatgen,
title = "{PERSONACHATGEN}: Generating Personalized Dialogues using {GPT}-3",
author = "Lee, Young-Jun and
Lim, Chae-Gyun and
Choi, Yunsu and
Lm, Ji-Hui and
Choi, Ho-Jin",
editor = "Lim, Heuiseok and
Kim, Seungryong and
Lee, Yeonsoo and
Lin, Steve and
Seo, Paul Hongsuck and
Suh, Yumin and
Jang, Yoonna and
Lim, Jungwoo and
Hur, Yuna and
Son, Suhyune",
booktitle = "Proceedings of the 1st Workshop on Customized Chat Grounding Persona and Knowledge",
month = oct,
year = "2022",
address = "Gyeongju, Republic of Korea",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.ccgpk-1.4",
pages = "29--48",
abstract = "Recently, many prior works have made their own agents generate more personalized and engaging responses using personachat. However, since this dataset is frozen in 2018, the dialogue agents trained on this dataset would not know how to interact with a human who loves {``}Wandavision.{''} One way to alleviate this problem is to create a large-scale dataset. In this work, we introduce the pipeline of creating personachatgen, which is comprised of three main components: Creating (1) profilegen, (2) Persona Set, and (3) personachatgen. To encourage GPT-3{'}s generation ability, we also defined a taxonomy of hierarchical persona category derived from social profiling taxonomy. To create the speaker consistent persona set, we propose a simple contradiction-based iterative sentence replacement algorithm, named CoNL. Moreover, to prevent GPT-3 generating harmful content, we presented two filtering pipelines, one each for profilegen and personachatgen. Through analyzing of personachatgen, we showed that GPT-3 can generate personalized dialogue containing diverse persona. Furthermore, we revealed a state-of-the-art Blender 90M trained on our dataset that leads to higher performance.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lee-etal-2022-personachatgen">
<titleInfo>
<title>PERSONACHATGEN: Generating Personalized Dialogues using GPT-3</title>
</titleInfo>
<name type="personal">
<namePart type="given">Young-Jun</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chae-Gyun</namePart>
<namePart type="family">Lim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yunsu</namePart>
<namePart type="family">Choi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ji-Hui</namePart>
<namePart type="family">Lm</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ho-Jin</namePart>
<namePart type="family">Choi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 1st Workshop on Customized Chat Grounding Persona and Knowledge</title>
</titleInfo>
<name type="personal">
<namePart type="given">Heuiseok</namePart>
<namePart type="family">Lim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Seungryong</namePart>
<namePart type="family">Kim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yeonsoo</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Steve</namePart>
<namePart type="family">Lin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Paul</namePart>
<namePart type="given">Hongsuck</namePart>
<namePart type="family">Seo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yumin</namePart>
<namePart type="family">Suh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yoonna</namePart>
<namePart type="family">Jang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jungwoo</namePart>
<namePart type="family">Lim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuna</namePart>
<namePart type="family">Hur</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Suhyune</namePart>
<namePart type="family">Son</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Gyeongju, Republic of Korea</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Recently, many prior works have made their own agents generate more personalized and engaging responses using personachat. However, since this dataset is frozen in 2018, the dialogue agents trained on this dataset would not know how to interact with a human who loves “Wandavision.” One way to alleviate this problem is to create a large-scale dataset. In this work, we introduce the pipeline of creating personachatgen, which is comprised of three main components: Creating (1) profilegen, (2) Persona Set, and (3) personachatgen. To encourage GPT-3’s generation ability, we also defined a taxonomy of hierarchical persona category derived from social profiling taxonomy. To create the speaker consistent persona set, we propose a simple contradiction-based iterative sentence replacement algorithm, named CoNL. Moreover, to prevent GPT-3 generating harmful content, we presented two filtering pipelines, one each for profilegen and personachatgen. Through analyzing of personachatgen, we showed that GPT-3 can generate personalized dialogue containing diverse persona. Furthermore, we revealed a state-of-the-art Blender 90M trained on our dataset that leads to higher performance.</abstract>
<identifier type="citekey">lee-etal-2022-personachatgen</identifier>
<location>
<url>https://aclanthology.org/2022.ccgpk-1.4</url>
</location>
<part>
<date>2022-10</date>
<extent unit="page">
<start>29</start>
<end>48</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T PERSONACHATGEN: Generating Personalized Dialogues using GPT-3
%A Lee, Young-Jun
%A Lim, Chae-Gyun
%A Choi, Yunsu
%A Lm, Ji-Hui
%A Choi, Ho-Jin
%Y Lim, Heuiseok
%Y Kim, Seungryong
%Y Lee, Yeonsoo
%Y Lin, Steve
%Y Seo, Paul Hongsuck
%Y Suh, Yumin
%Y Jang, Yoonna
%Y Lim, Jungwoo
%Y Hur, Yuna
%Y Son, Suhyune
%S Proceedings of the 1st Workshop on Customized Chat Grounding Persona and Knowledge
%D 2022
%8 October
%I Association for Computational Linguistics
%C Gyeongju, Republic of Korea
%F lee-etal-2022-personachatgen
%X Recently, many prior works have made their own agents generate more personalized and engaging responses using personachat. However, since this dataset is frozen in 2018, the dialogue agents trained on this dataset would not know how to interact with a human who loves “Wandavision.” One way to alleviate this problem is to create a large-scale dataset. In this work, we introduce the pipeline of creating personachatgen, which is comprised of three main components: Creating (1) profilegen, (2) Persona Set, and (3) personachatgen. To encourage GPT-3’s generation ability, we also defined a taxonomy of hierarchical persona category derived from social profiling taxonomy. To create the speaker consistent persona set, we propose a simple contradiction-based iterative sentence replacement algorithm, named CoNL. Moreover, to prevent GPT-3 generating harmful content, we presented two filtering pipelines, one each for profilegen and personachatgen. Through analyzing of personachatgen, we showed that GPT-3 can generate personalized dialogue containing diverse persona. Furthermore, we revealed a state-of-the-art Blender 90M trained on our dataset that leads to higher performance.
%U https://aclanthology.org/2022.ccgpk-1.4
%P 29-48
Markdown (Informal)
[PERSONACHATGEN: Generating Personalized Dialogues using GPT-3](https://aclanthology.org/2022.ccgpk-1.4) (Lee et al., CCGPK 2022)
ACL
- Young-Jun Lee, Chae-Gyun Lim, Yunsu Choi, Ji-Hui Lm, and Ho-Jin Choi. 2022. PERSONACHATGEN: Generating Personalized Dialogues using GPT-3. In Proceedings of the 1st Workshop on Customized Chat Grounding Persona and Knowledge, pages 29–48, Gyeongju, Republic of Korea. Association for Computational Linguistics.