Casevo: A Cognitive Agents and Social Evolution Simulator

2024-12-27

AI Summary

Traditional agent-based modeling (ABM) struggles to capture individual cognitive evolution and dynamic social interaction. To address this, we propose the first method embedding large language models (LLMs) deeply into a discrete-event-driven ABM framework. Our approach integrates chain-of-thought reasoning (CoT), retrieval-augmented generation (RAG), and a customizable memory mechanism—enabling cognitively interpretable, memory-sustaining, and interaction-adaptive social agents. Unlike static rule-based ABM, our framework significantly enhances behavioral authenticity and the dynamic representational fidelity of social processes. We validate it on the 2020 U.S. midterm election televised debate scenario, demonstrating high-fidelity simulation of opinion dynamics, networked information diffusion, and ideological polarization. The framework advances computational social science by offering a novel paradigm that balances interpretability with modeling flexibility.

Technology Category

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

Application Category

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

Abstract

In this paper, we introduce a multi-agent simulation framework Casevo (Cognitive Agents and Social Evolution Simulator), that integrates large language models (LLMs) to simulate complex social phenomena and decision-making processes. Casevo is designed as a discrete-event simulator driven by agents with features such as Chain of Thoughts (CoT), Retrieval-Augmented Generation (RAG), and Customizable Memory Mechanism. Casevo enables dynamic social modeling, which can support various scenarios such as social network analysis, public opinion dynamics, and behavior prediction in complex social systems. To demonstrate the effectiveness of Casevo, we utilize one of the U.S. 2020 midterm election TV debates as a simulation example. Our results show that Casevo facilitates more realistic and flexible agent interactions, improving the quality of dynamic social phenomena simulation. This work contributes to the field by providing a robust system for studying large-scale, high-fidelity social behaviors with advanced LLM-driven agents, expanding the capabilities of traditional agent-based modeling (ABM). The open-source code repository address of casevo is https://github.com/rgCASS/casevo.

Problem

• Social Dynamics
• Behavior Prediction
• Complex Event Analysis

Innovation

• Casevo
• Social Behavior Simulation
• Opinion Evolution