OASIS: Open Agent Social Interaction Simulations with One Million Agents

2024-11-18

arXiv.org

AI Summary

Existing LLM-driven agent-based models (ABMs) suffer from scenario-specific design and scalability limitations, hindering large-scale simulation of social media systems with millions of users. This paper introduces the first scalable, general-purpose social platform simulation framework enabling LLM agents to co-evolve at scale within dynamic graph networks, multi-action spaces, and a dual-mode recommendation mechanism—integrating both interest-based and popularity-based signals—while leveraging a distributed simulation engine. The framework supports cross-platform adaptation and multi-scale observation of social phenomena, overcoming traditional LLM-ABM constraints in both scalability (supporting up to million-scale agents) and generalizability. Experiments successfully reproduce canonical social dynamics—including information diffusion, group polarization, and herding behavior—demonstrating that scaling significantly enhances diversity in collective dynamics and improves opinion quality.

Technology Category

• Multiagent Systems: Agent-Based Simulation and Emergent Behavior
• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• 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

Abstract

There has been a growing interest in enhancing rule-based agent-based models (ABMs) for social media platforms (i.e., X, Reddit) with more realistic large language model (LLM) agents, thereby allowing for a more nuanced study of complex systems. As a result, several LLM-based ABMs have been proposed in the past year. While they hold promise, each simulator is specifically designed to study a particular scenario, making it time-consuming and resource-intensive to explore other phenomena using the same ABM. Additionally, these models simulate only a limited number of agents, whereas real-world social media platforms involve millions of users. To this end, we propose OASIS, a generalizable and scalable social media simulator. OASIS is designed based on real-world social media platforms, incorporating dynamically updated environments (i.e., dynamic social networks and post information), diverse action spaces (i.e., following, commenting), and recommendation systems (i.e., interest-based and hot-score-based). Additionally, OASIS supports large-scale user simulations, capable of modeling up to one million users. With these features, OASIS can be easily extended to different social media platforms to study large-scale group phenomena and behaviors. We replicate various social phenomena, including information spreading, group polarization, and herd effects across X and Reddit platforms. Moreover, we provide observations of social phenomena at different agent group scales. We observe that the larger agent group scale leads to more enhanced group dynamics and more diverse and helpful agents' opinions. These findings demonstrate OASIS's potential as a powerful tool for studying complex systems in digital environments.

Problem

• Enhancing rule-based ABMs with realistic LLM agents
• Creating scalable simulator for large-scale social phenomena
• Studying diverse group dynamics across platforms

Innovation

• Generalizable social media simulator with dynamic environments
• Scalable model supporting one million LLM agents
• Diverse action spaces and recommendation systems integration