GASim: A Graph-Accelerated Hybrid Framework for Social Simulation

2026-05-08

AI Summary

This work addresses the high computational overhead of memory retrieval in large language models (LLMs) and the latency induced by serial execution in agent-based modeling (ABM) for large-scale social simulations. To overcome these limitations, the authors propose a graph-accelerated hybrid multi-agent framework that innovatively integrates graph neural networks with social simulation. The framework introduces three core mechanisms: Graph-Optimized Memory (GOM), Graph Message Passing (GMP), and Entropy-Driven Grouping (EDG), enabling efficient parallel computation and dynamic hierarchical organization of agents. Experimental results demonstrate that the system achieves a 9.94× end-to-end speedup and reduces token consumption by 20%, while maintaining high fidelity to real-world opinion dynamics, thereby significantly enhancing both simulation efficiency and scalability.

Abstract

Large-scale social simulators are essential for studying complex social patterns. Prior work explores hybrid methods to scale up simulations, combining large language models (LLM)-based agents with numerical agent-based models (ABM). However, this incurs high latency due to expensive memory retrieval and sequential ABM execution. To address this challenge, we propose GASim, a graph-accelerated hybrid multi-agent framework for large-scale social simulations. For core agents driven by LLM, GASim introduces Graph-Optimized Memory (GOM) to replace intensive LLM-based retrieval pipelines with lightweight propagation over a sparse memory graph. For the majority of ordinary agents, GASim employs Graph Message Passing (GMP), substituting sequential ABM execution with parallel updates by fine-grained feature aggregation and Graph Attention Network. We further introduce Entropy-Driven Grouping (EDG) that coordinates this hybrid partitioning, leveraging information entropy to dynamically identify emergent core agents situated in information-diverse neighborhoods. Extensive experiments show that GASim not only delivers a substantial 9.94-fold end-to-end speedup over the traditional hybrid framework but also consumes less than 20% of baseline tokens, significantly reducing costs while preserving strong alignment with real-world public opinion trends. Our code is available at https://github.com/Jasmine0201/GASim.

Problem

• social simulation
• hybrid framework
• large language models
• agent-based models
• latency

Innovation

• Graph-Optimized Memory
• Graph Message Passing
• Entropy-Driven Grouping
• Hybrid Multi-Agent Simulation
• Large Language Models