Attention Mechanism for LLM-based Agents Dynamic Diffusion under Information Asymmetry

TL;DR

This paper introduces a dynamic attention mechanism for large language model-based multi-agent systems to better simulate social relationships, information diffusion, and social capital development in asymmetric environments.

Contribution

It proposes a novel framework and attention mechanism to improve LLM agents' perception of social relationships and information diffusion in complex, asymmetric settings.

Findings

Agents form information circles and develop relationships.

Information gaps and diffusion patterns evolve over time.

Social capital accumulation is observed in the simulation.

Abstract

Large language models have been used to simulate human society using multi-agent systems. Most current social simulation research emphasizes interactive behaviors in fixed environments, ignoring information opacity, relationship variability, and diffusion diversity. In this paper, we first propose a general framework for exploring multi-agent information diffusion. We identified LLMs' deficiency in the perception and utilization of social relationships, as well as diverse actions. Then, we designed a dynamic attention mechanism to help agents allocate attention to different information, addressing the limitations of the LLM attention mechanism. Agents start by responding to external information stimuli within a five-agent group, increasing group size and forming information circles while developing relationships and sharing information. Additionally, we explore the information diffusion features in the asymmetric open environment by observing the evolution of information gaps, diffusion patterns, and the accumulation of social capital, which are closely linked to psychological, sociological, and communication theories.