@inproceedings{wang-etal-2024-sotopia,
title = "{SOTOPIA}-{\mbox{$\pi$}}: Interactive Learning of Socially Intelligent Language Agents",
author = "Wang, Ruiyi and
Yu, Haofei and
Zhang, Wenxin and
Qi, Zhengyang and
Sap, Maarten and
Bisk, Yonatan and
Neubig, Graham and
Zhu, Hao",
editor = "Ku, Lun-Wei and
Martins, Andre and
Srikumar, Vivek",
booktitle = "Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = aug,
year = "2024",
address = "Bangkok, Thailand",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.acl-long.698",
doi = "10.18653/v1/2024.acl-long.698",
pages = "12912--12940",
abstract = "Humans learn social skills through both imitation and social interaction. This social learning process is largely understudied by existing research on building language agents. Motivated by this gap, we propose an interactive learning method, SOTOPIA-{\mbox{$\pi$}}, that improves the social intelligence of language agents. This method leverages behavior cloning and self-reinforcement based training on filtered social interaction data according to large language model (LLM) rating. We show that our training method allows a 7B LLM to reach the social goal completion ability of an expert model (GPT-4-based agent) without the loss of more generic abilities, such as the ability to answer knowledge-based questions. We also demonstrate that this training paradigm uncovers some weaknesses in standard evaluation and safety training paradigms that (1) LLM-based evaluation of social intelligence overestimates the abilities of the language agents trained specifically for social interaction, and that (2) despite not training for better safety or question answering (QA) ability, our methods improve the safety of language agents and maintain general QA ability on the MMLU benchmark.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2024-sotopia">
<titleInfo>
<title>SOTOPIA-π: Interactive Learning of Socially Intelligent Language Agents</title>
</titleInfo>
<name type="personal">
<namePart type="given">Ruiyi</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haofei</namePart>
<namePart type="family">Yu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wenxin</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhengyang</namePart>
<namePart type="family">Qi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maarten</namePart>
<namePart type="family">Sap</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yonatan</namePart>
<namePart type="family">Bisk</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Graham</namePart>
<namePart type="family">Neubig</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hao</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lun-Wei</namePart>
<namePart type="family">Ku</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andre</namePart>
<namePart type="family">Martins</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vivek</namePart>
<namePart type="family">Srikumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Bangkok, Thailand</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Humans learn social skills through both imitation and social interaction. This social learning process is largely understudied by existing research on building language agents. Motivated by this gap, we propose an interactive learning method, SOTOPIA-π, that improves the social intelligence of language agents. This method leverages behavior cloning and self-reinforcement based training on filtered social interaction data according to large language model (LLM) rating. We show that our training method allows a 7B LLM to reach the social goal completion ability of an expert model (GPT-4-based agent) without the loss of more generic abilities, such as the ability to answer knowledge-based questions. We also demonstrate that this training paradigm uncovers some weaknesses in standard evaluation and safety training paradigms that (1) LLM-based evaluation of social intelligence overestimates the abilities of the language agents trained specifically for social interaction, and that (2) despite not training for better safety or question answering (QA) ability, our methods improve the safety of language agents and maintain general QA ability on the MMLU benchmark.</abstract>
<identifier type="citekey">wang-etal-2024-sotopia</identifier>
<identifier type="doi">10.18653/v1/2024.acl-long.698</identifier>
<location>
<url>https://aclanthology.org/2024.acl-long.698</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>12912</start>
<end>12940</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T SOTOPIA-π: Interactive Learning of Socially Intelligent Language Agents
%A Wang, Ruiyi
%A Yu, Haofei
%A Zhang, Wenxin
%A Qi, Zhengyang
%A Sap, Maarten
%A Bisk, Yonatan
%A Neubig, Graham
%A Zhu, Hao
%Y Ku, Lun-Wei
%Y Martins, Andre
%Y Srikumar, Vivek
%S Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand
%F wang-etal-2024-sotopia
%X Humans learn social skills through both imitation and social interaction. This social learning process is largely understudied by existing research on building language agents. Motivated by this gap, we propose an interactive learning method, SOTOPIA-π, that improves the social intelligence of language agents. This method leverages behavior cloning and self-reinforcement based training on filtered social interaction data according to large language model (LLM) rating. We show that our training method allows a 7B LLM to reach the social goal completion ability of an expert model (GPT-4-based agent) without the loss of more generic abilities, such as the ability to answer knowledge-based questions. We also demonstrate that this training paradigm uncovers some weaknesses in standard evaluation and safety training paradigms that (1) LLM-based evaluation of social intelligence overestimates the abilities of the language agents trained specifically for social interaction, and that (2) despite not training for better safety or question answering (QA) ability, our methods improve the safety of language agents and maintain general QA ability on the MMLU benchmark.
%R 10.18653/v1/2024.acl-long.698
%U https://aclanthology.org/2024.acl-long.698
%U https://doi.org/10.18653/v1/2024.acl-long.698
%P 12912-12940
Markdown (Informal)
[SOTOPIA-π: Interactive Learning of Socially Intelligent Language Agents](https://aclanthology.org/2024.acl-long.698) (Wang et al., ACL 2024)
ACL
- Ruiyi Wang, Haofei Yu, Wenxin Zhang, Zhengyang Qi, Maarten Sap, Yonatan Bisk, Graham Neubig, and Hao Zhu. 2024. SOTOPIA-π: Interactive Learning of Socially Intelligent Language Agents. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 12912–12940, Bangkok, Thailand. Association for Computational Linguistics.