SocioVerse: A World Model for Social Simulation Powered by LLM Agents and A Pool of 10 Million Real-World Users

2025-04-14

AI Summary

Existing social simulation approaches suffer from misalignment with real-world societies across four critical dimensions—environment, users, interactions, and behaviors—resulting in inadequate modeling of individual heterogeneity and inaccurate prediction of collective dynamics. To address this, we propose the first large language model (LLM)-driven world model explicitly designed for social simulation, establishing a novel “Environment–User–Interaction–Behavior” (EUIB) alignment framework. Our method integrates LLM-based intelligent agents, multi-scale behavioral modeling, standardized simulation protocols, and a real-data co-training mechanism, leveraging millions of authentic user records to enable large-scale, high-fidelity, and generalizable social dynamic simulation. Evaluated across political, news, and economic domains, the model accurately reproduces population-level evolutionary patterns while preserving behavioral diversity, systemic plausibility, and statistical representativeness; human intervention is required in fewer than 5% of cases.

Technology Category

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

Application Category

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

Abstract

Social simulation is transforming traditional social science research by modeling human behavior through interactions between virtual individuals and their environments. With recent advances in large language models (LLMs), this approach has shown growing potential in capturing individual differences and predicting group behaviors. However, existing methods face alignment challenges related to the environment, target users, interaction mechanisms, and behavioral patterns. To this end, we introduce SocioVerse, an LLM-agent-driven world model for social simulation. Our framework features four powerful alignment components and a user pool of 10 million real individuals. To validate its effectiveness, we conducted large-scale simulation experiments across three distinct domains: politics, news, and economics. Results demonstrate that SocioVerse can reflect large-scale population dynamics while ensuring diversity, credibility, and representativeness through standardized procedures and minimal manual adjustments.

Problem

• Aligning virtual environments with real-world user behaviors
• Capturing individual differences in large-scale social simulations
• Ensuring diversity and credibility in simulated population dynamics

Innovation

• LLM-agent-driven world model for social simulation
• Four powerful alignment components for accuracy
• 10 million real users pool for representativeness