Go to ACL ARR 2024 June homepageWar and Peace (WarAgent): LLM-based Multi-Agent Simulation of World Wars
Abstract: Can we avoid wars at the crossroads of history? This question has been pursued by scholars, policymakers, and organizations throughout human history. In this research, we attempt to answer the question based on the recent advances of Artificial Intelligence and Large Language Models. We propose WarAgent, an LLM-powered multi-agent AI system, to simulate the participating countries, their decisions, and the consequences, in three historical international conflicts. Based on WarAgent, we also propose standard evaluation protocols for LLM-based multi-agent simulation systems, which helps the community to advance on the important evaluation problem of multi-agent systems. By evaluating the simulation effectiveness, we examine the advancements and limitations of cutting-edge AI systems' abilities in studying complex collective human behaviors such as international conflicts under diverse settings. In these simulations, the emergent interactions among agents also offer a novel perspective for examining the triggers and conditions that lead to war. Our findings offer data-driven and AI-augmented insights that can redefine how we approach conflict resolution and peacekeeping strategies. The implications stretch beyond computer simulation and historical analysis, offering a blueprint for using AI to understand human history and possibly prevent future international conflicts. Code and data are available at \url{https://anonymous.4open.science/r/war_and_peace-061E}.
Paper Type: Long
Research Area: Computational Social Science and Cultural Analytics
Research Area Keywords: large language model, agent, multi-agent system, social simulation
Contribution Types: NLP engineering experiment
Languages Studied: English
Submission Number: 3472
Loading