CaM-Wolf: Causal-Aware Multimodal Agents for Social Deduction Games

• URL: http://arxiv.org/abs/2607.26393v1
• Date: Wed, 29 Jul 2026 01:58:36 GMT
• Title: CaM-Wolf: Causal-Aware Multimodal Agents for Social Deduction Games
• Abstract summary: Social deduction games (SDGs) such as Werewolf have become challenging testbeds for AI agents.<n>CaM-Wolf processes video inputs from other players, employs a causal-aware Reasoner trained via reinforcement learning to establish logical chains between observable behaviors and hidden roles.<n>Our experiments and user study show that CaM-Wolf achieves superior agent gameplay performance and enhances the quality of human-AI interaction.
• Score: 41.77561512638328
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Social deduction games (SDGs) such as Werewolf have become challenging testbeds for AI agents. These games require complex social skills such as reasoning, deception, and collaboration. While recent advances in large language models (LLMs) have driven significant progress in SDG agents, current approaches are predominantly text-based, overlooking the multimodal nature that is fundamental to human social interaction. To bridge this gap, we introduce CaM-Wolf, the first SDG agent that integrates multimodal perception and generation. CaM-Wolf processes video inputs from other players, employs a causal-aware Reasoner trained via reinforcement learning to establish logical chains between observable behaviors and hidden roles, and presents itself through an animated avatar. Our experiments and user study show that CaM-Wolf achieves superior agent gameplay performance and enhances the quality of human-AI interaction. This work represents a significant advancement towards creating more human-like AI agents capable of participating in nuanced social dynamics. Our code is available at https://3dagentworld.github.io/avatar_wolf.

Related papers

• Enhancing Consistency of Werewolf AI through Dialogue Summarization and Persona Information [2.3101858435998346]
  Werewolf Game is a communication game where players' reasoning and discussion skills are essential.<n>We present a Werewolf AI agent developed for the AIWolfDial 2024 shared task, co-hosted with the 17th INLG.
  arXiv  Detail & Related papers  (2026-03-07T08:41:46Z)
• HumanX: Toward Agile and Generalizable Humanoid Interaction Skills from Human Videos [74.43500240121476]
  We present HumanX, a full-stack framework that compiles human video into generalizable, real-world interaction skills for humanoids.<n>HumanX achieves over 8 times higher generalization success than prior methods.
  arXiv  Detail & Related papers  (2026-02-02T18:53:01Z)
• Verbal Werewolf: Engage Users with Verbalized Agentic Werewolf Game Framework [40.58507305270978]
  Social deduction games like Werewolf, traditionally played through verbal communication, present an ideal application for Large Language Models (LLMs)<n>We propose textbfVerbal Werewolf, a novel LLM-based game system that optimize two parallel pipelines: gameplay powered by state-of-the-art LLMs and a fine-tuned Text-to-Speech (TTS) module that brings text output to life.
  arXiv  Detail & Related papers  (2025-05-30T18:58:57Z)
• MultiMind: Enhancing Werewolf Agents with Multimodal Reasoning and Theory of Mind [19.859451521261956]
  MultiMind is the first framework integrating multimodal information into social deduction agents.<n>It processes facial expressions and vocal tones alongside verbal content, while employing a Theory of Mind (ToM) model.<n>By combining this ToM model with Monte Carlo Tree Search (MCTS), our agent identifies communication strategies that minimize suspicion directed at itself.
  arXiv  Detail & Related papers  (2025-04-25T03:12:43Z)
• Towards Anthropomorphic Conversational AI Part I: A Practical Framework [49.62013440962072]
  We introduce a multi- module framework designed to replicate the key aspects of human intelligence involved in conversations.<n>In the second stage of our approach, these conversational data, after filtering and labeling, can serve as training and testing data for reinforcement learning.
  arXiv  Detail & Related papers  (2025-02-28T03:18:39Z)
• SocialGFs: Learning Social Gradient Fields for Multi-Agent Reinforcement Learning [58.84311336011451]
  We propose a novel gradient-based state representation for multi-agent reinforcement learning. We employ denoising score matching to learn the social gradient fields (SocialGFs) from offline samples. In practice, we integrate SocialGFs into the widely used multi-agent reinforcement learning algorithms, e.g., MAPPO.
  arXiv  Detail & Related papers  (2024-05-03T04:12:19Z)
• Toward Human-AI Alignment in Large-Scale Multi-Player Games [24.784173202415687]
  We analyze extensive human gameplay data from Xbox's Bleeding Edge (100K+ games) We find that while human players exhibit variability in fight-flight and explore-exploit behavior, AI players tend towards uniformity. These stark differences underscore the need for interpretable evaluation, design, and integration of AI in human-aligned applications.
  arXiv  Detail & Related papers  (2024-02-05T22:55:33Z)
• The Rise and Potential of Large Language Model Based Agents: A Survey [91.71061158000953]
  Large language models (LLMs) are regarded as potential sparks for Artificial General Intelligence (AGI) We start by tracing the concept of agents from its philosophical origins to its development in AI, and explain why LLMs are suitable foundations for agents. We explore the extensive applications of LLM-based agents in three aspects: single-agent scenarios, multi-agent scenarios, and human-agent cooperation.
  arXiv  Detail & Related papers  (2023-09-14T17:12:03Z)
• Playing the Werewolf game with artificial intelligence for language understanding [0.7550566004119156]
  Werewolf is a social deduction game based on free natural language communication. The purpose of this study is to develop an AI agent that can play Werewolf through natural language conversations.
  arXiv  Detail & Related papers  (2023-02-21T13:03:20Z)
• Incorporating Rivalry in Reinforcement Learning for a Competitive Game [65.2200847818153]
  This work proposes a novel reinforcement learning mechanism based on the social impact of rivalry behavior. Our proposed model aggregates objective and social perception mechanisms to derive a rivalry score that is used to modulate the learning of artificial agents.
  arXiv  Detail & Related papers  (2022-08-22T14:06:06Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.