LLM-Based Agent Society Investigation: Collaboration and Confrontation in Avalon Gameplay
This addresses the open research problem of social behaviors in LLM agents for AI and multi-agent systems research, though it is incremental as it builds on existing gameplay studies.
The paper tackled the problem of understanding social behaviors in LLM-based agents by using Avalon gameplay as a testbed, resulting in a framework that effectively creates adaptive agents and demonstrates their potential in dynamic social interactions.
This paper explores the open research problem of understanding the social behaviors of LLM-based agents. Using Avalon as a testbed, we employ system prompts to guide LLM agents in gameplay. While previous studies have touched on gameplay with LLM agents, research on their social behaviors is lacking. We propose a novel framework, tailored for Avalon, features a multi-agent system facilitating efficient communication and interaction. We evaluate its performance based on game success and analyze LLM agents' social behaviors. Results affirm the framework's effectiveness in creating adaptive agents and suggest LLM-based agents' potential in navigating dynamic social interactions. By examining collaboration and confrontation behaviors, we offer insights into this field's research and applications. Our code is publicly available at https://github.com/3DAgentWorld/LLM-Game-Agent.