DVM: Towards Controllable LLM Agents In Social Deduction Games

Abstract

Large Language Models (LLMs) have advanced the capability of game agents in social deduction games (SDGs). These games rely heavily on conversation-driven interactions and require agents to infer, make decisions, and express based on such information. While this progress leads to more sophisticated and strategic non-player characters (NPCs) in SDGs, there exists a need to control the proficiency of these agents. This control not only ensures that NPCs can adapt to varying difficulty levels during gameplay, but also provides insights into the safety and fairness of LLM agents. In this paper, we present DVM, a novel framework for developing controllable LLM agents for SDGs, and demonstrate its implementation on one of the most popular SDGs, Werewolf. DVM comprises three main components: Predictor, Decider, and Discussor. By integrating reinforcement learning with a win rate-constrained decision chain reward mechanism, we enable agents to dynamically adjust their gameplay proficiency to a

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyzhang2025dvm

Related papers

• Language Agents With Reinforcement Learning For Strategic Play In The Werewolf Game (2023)0.00
• DLM: Unified Decision Language Models For Offline Multi-agent Sequential Decision Making (2026)0.00
• Reinforcement Learning Environment With Llm-controlled Adversary In D&D 5th Edition Combat (2025)0.00
• Rlupus: Cooperation Through Emergent Communication In The Werewolf Social Deduction Game (2021)0.00
• Enhancing Vision-language Model Training With Reinforcement Learning In Synthetic Worlds For Real-world Success (2025)0.00
• Language-driven Coordination And Learning In Multi-agent Simulation Environments (2025)0.00
• SAC-GLAM: Improving Online RL For LLM Agents With Soft Actor-critic And Hindsight Relabeling (2024)0.00
• Agent-pro: Learning To Evolve Via Policy-level Reflection And Optimization (2024)9.59