AMONGAGENTS: Evaluating Large Language Models in the Interactive Text-Based Social Deduction Game
This work addresses the problem of assessing LLM performance in socially driven, goal-oriented games with incomplete information for researchers in AI and social science, though it is incremental as it applies existing methods to a new domain.
The paper tackled evaluating large language models (LLMs) in interactive social deduction games by creating AmongAgents, a text-based simulation of Among Us, and found that state-of-the-art LLMs can effectively understand game rules and make context-based decisions.
Strategic social deduction games serve as valuable testbeds for evaluating the understanding and inference skills of language models, offering crucial insights into social science, artificial intelligence, and strategic gaming. This paper focuses on creating proxies of human behavior in simulated environments, with Among Us utilized as a tool for studying simulated human behavior. The study introduces a text-based game environment, named AmongAgents, that mirrors the dynamics of Among Us. Players act as crew members aboard a spaceship, tasked with identifying impostors who are sabotaging the ship and eliminating the crew. Within this environment, the behavior of simulated language agents is analyzed. The experiments involve diverse game sequences featuring different configurations of Crewmates and Impostor personality archetypes. Our work demonstrates that state-of-the-art large language models (LLMs) can effectively grasp the game rules and make decisions based on the current context. This work aims to promote further exploration of LLMs in goal-oriented games with incomplete information and complex action spaces, as these settings offer valuable opportunities to assess language model performance in socially driven scenarios.