Learning to Discuss Strategically: A Case Study on One Night Ultimate Werewolf
Communication is a fundamental aspect of human society, facilitating the exchange of information and beliefs among people. Despite the advancements in large language models (LLMs), recent agents built with these often neglect the control over discussion tactics, which are essential in communication scenarios and games. As a variant of the famous communication game Werewolf, One Night Ultimate Werewolf (ONUW) requires players to develop strategic discussion policies due to the potential role changes that increase the uncertainty and complexity of the game. In this work, we first present the existence of the Perfect Bayesian Equilibria (PBEs) in two scenarios of the ONUW game: one with discussion and one without. The results showcase that the discussion greatly changes players' utilities by affecting their beliefs, emphasizing the significance of discussion tactics. Based on the insights obtained from the analyses, we propose an RL-instructed language agent framework, where a discussion policy trained by reinforcement learning (RL) is employed to determine appropriate discussion tactics to adopt. Our experimental results on several ONUW game settings demonstrate the effectiveness and generalizability of our proposed framework. The project page of our paper: $\href{https://one-night-ultimate-werewolf.github.io}{one-night-ultimate-werewolf.github.io}$.
Code (0)
등록된 구현이 없습니다.
Tasks
Reinforcement Learning (RL)Similar Papers 제목 키워드 기반
Disentangled Contrastive Image Translation for Nighttime Surveillance
Nighttime surveillance suffers from degradation due to poor illumination and arduous human annotations. It is challengable and remains a security risk at night. Existing methods rely on multi-spectral images to perceive …
Contrastive LearningTranslationBack-filling Missing Data When Predicting Domestic Electricity Consumption From Smart Meter Data
This study uses data from domestic electricity smart meters to estimate annual electricity bills for a whole year. We develop a method for back-filling data smart meter for up to six missing months for users who have les…
Forecasting of a Hierarchical Functional Time Series on Example of Macromodel for Day and Night Air Pollution in Silesia Region: A Critical Overview
In economics we often face a system, which intrinsically imposes a structure of hierarchy of its components, i.e., in modelling trade accounts related to foreign exchange or in optimization of regional air protection pol…
Time SeriesTime Series AnalysisArbitrage-free modeling under Knightian Uncertainty
We study the Fundamental Theorem of Asset Pricing for a general financial market under Knightian Uncertainty. We adopt a functional analytic approach which require neither specific assumptions on the class of priors $\ma…
Werewolf Arena: A Case Study in LLM Evaluation via Social Deduction
This paper introduces Werewolf Arena, a novel framework for evaluating large language models (LLMs) through the lens of the classic social deduction game, Werewolf. In Werewolf Arena, LLMs compete against each other, nav…