arXiv cs.CL· Yoshiki Tanaka, Shuya Uchiyama, Daichi Hamaguchi, Michimasa Inaba·· 12 小时前AI 评分20
狼人杀 AI 智能体 Adversarial Inner Thoughts：用内心思考决定何时发言
Beyond What to Say: Proactive Utterance Timing with Inner Thoughts for Werewolf Agents
AI 导读
针对 AIWolfDial 共享任务 Anytime Utterance Track，研究者提出 Adversarial Inner Thoughts 方法，在 Inner Thoughts 框架基础上为狼人杀这类利益冲突、信息不完整的场景重新设计发言时机判断。
正文 · 原文
Abstract:Participating in a conversation involves not only deciding what to say but also whether to speak at all. This judgment carries even greater significance in environments like the Werewolf game, where participants' interests conflict, because both utterances and silence are subject to interpretation and assessment by others. Therefore, determining when to speak, as well as what to say, is crucial for an LLM agent playing the Werewolf game. In this study, we introduce a Werewolf AI agent developed for the Anytime Utterance Track of the AIWolfDial shared task. Our system builds on the existing Inner Thoughts framework, which determines utterance timing during conversations by generating internal thoughts and evaluating the value of each thought for expression at that moment. Since existing methods were designed for general dialogue, we have redesigned the system specifically for the competitive environment and incomplete information setting of Werewolf games. We call our adapted method Adversarial Inner Thoughts. Through experiments involving analysis of self-play logs and human evaluations, we analyze and discuss the effectiveness of our approach.
| Comments: | Accepted to the 4th International AIWolfDial Workshop at INLG 2026 |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.32107 [cs.CL] |
| (or arXiv:2610.32107v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.32107 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yoshiki Tanaka [view email]
[v1]
Fri, 23 Oct 2026 03:19:33 UTC (612 KB)
来源：arXiv cs.CL · arxiv.org