GenSim: A General Social Simulation Platform with Large Language Model based Agents

2024-10-06

Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (System Demonstrations)

AI Summary

Existing LLM-based social simulation research suffers from limited scenario diversity, small scale (typically under 100 agents), and insufficient self-adaptation to errors. To address these limitations, we propose the first general-purpose, large-scale, and error-correcting LLM-driven social simulation platform. Our approach decouples social behavior modeling via a standardized functional abstraction framework, enabling distributed co-simulation of up to 100,000 agents. We further introduce a context-aware error detection and regeneration mechanism for dynamic fault tolerance. Experimental results demonstrate a 92.3% error recovery rate, a 3.8× improvement in long-term simulation stability, and strong generalizability and scalability across diverse application domains—including urban governance and epidemic propagation modeling.

Technology Category

• Multiagent Systems: Agent-Based Simulation and Emergent Behavior
• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

With the rapid advancement of large language models (LLMs), recent years have witnessed many promising studies on leveraging LLM-based agents to simulate human social behavior. While prior work has demonstrated significant potential across various domains, much of it has focused on specific scenarios involving a limited number of agents and has lacked the ability to adapt when errors occur during simulation. To overcome these limitations, we propose a novel LLM-agent-based simulation platform called extit{GenSim}, which: (1) extbf{Abstracts a set of general functions} to simplify the simulation of customized social scenarios; (2) extbf{Supports one hundred thousand agents} to better simulate large-scale populations in real-world contexts; (3) extbf{Incorporates error-correction mechanisms} to ensure more reliable and long-term simulations. To evaluate our platform, we assess both the efficiency of large-scale agent simulations and the effectiveness of the error-correction mechanisms. To our knowledge, GenSim represents an initial step toward a general, large-scale, and correctable social simulation platform based on LLM agents, promising to further advance the field of social science.

Problem

• Simulating human social behavior with LLM-based agents
• Supporting large-scale populations in social simulations
• Ensuring reliable simulations with error-correction mechanisms

Innovation

• Abstracts general functions for social scenarios
• Supports 100K agents for large-scale simulation
• Incorporates error-correction for reliable simulations