Enhancing Consistency of Werewolf AI through Dialogue Summarization and Persona Information

Yoshiki Tanaka, Takumasa Kaneko, Hiroki Onozeki, Natsumi Ezure, Ryuichi Uehara, Zhiyang Qi, Tomoya Higuchi, Ryutaro Asahara, Michimasa Inaba


Abstract
The Werewolf Game is a communication game where players’ reasoning and discussion skills are essential. In this study, we present a Werewolf AI agent developed for the AIWolfDial 2024 shared task, co-hosted with the 17th INLG. In recent years, large language models like ChatGPT have garnered attention for their exceptional response generation and reasoning capabilities. We thus develop the LLM-based agents for the Werewolf Game. This study aims to enhance the consistency of the agent’s utterances by utilizing dialogue summaries generated by LLMs and manually designed personas and utterance examples. By analyzing self-match game logs, we demonstrate that the agent’s utterances are contextually consistent and that the character, including tone, is maintained throughout the game.
Anthology ID:
2024.aiwolfdial-1.6
Volume:
Proceedings of the 2nd International AIWolfDial Workshop
Month:
September
Year:
2024
Address:
Tokyo, Japan
Editor:
Yoshinobu Kano
Venues:
AIWolfDial | WS
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
48–57
Language:
URL:
https://aclanthology.org/2024.aiwolfdial-1.6
DOI:
Bibkey:
Cite (ACL):
Yoshiki Tanaka, Takumasa Kaneko, Hiroki Onozeki, Natsumi Ezure, Ryuichi Uehara, Zhiyang Qi, Tomoya Higuchi, Ryutaro Asahara, and Michimasa Inaba. 2024. Enhancing Consistency of Werewolf AI through Dialogue Summarization and Persona Information. In Proceedings of the 2nd International AIWolfDial Workshop, pages 48–57, Tokyo, Japan. Association for Computational Linguistics.
Cite (Informal):
Enhancing Consistency of Werewolf AI through Dialogue Summarization and Persona Information (Tanaka et al., AIWolfDial-WS 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.aiwolfdial-1.6.pdf