Towards Simulating Social Influence Dynamics with LLM-based Multi-agents

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Lin, Hsien-Tsung, Huang, Pei-Cing, Ku, Chan-Tung, Hsu, Chan, Shieh, Pei-Xuan, Kang, Yihuang
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909712209936384
author Lin, Hsien-Tsung
Huang, Pei-Cing
Ku, Chan-Tung
Hsu, Chan
Shieh, Pei-Xuan
Kang, Yihuang
author_facet Lin, Hsien-Tsung
Huang, Pei-Cing
Ku, Chan-Tung
Hsu, Chan
Shieh, Pei-Xuan
Kang, Yihuang
contents Recent advancements in Large Language Models offer promising capabilities to simulate complex human social interactions. We investigate whether LLM-based multi-agent simulations can reproduce core human social dynamics observed in online forums. We evaluate conformity dynamics, group polarization, and fragmentation across different model scales and reasoning capabilities using a structured simulation framework. Our findings indicate that smaller models exhibit higher conformity rates, whereas models optimized for reasoning are more resistant to social influence.
format Preprint
id arxiv_https___arxiv_org_abs_2507_22467
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Towards Simulating Social Influence Dynamics with LLM-based Multi-agents
Lin, Hsien-Tsung
Huang, Pei-Cing
Ku, Chan-Tung
Hsu, Chan
Shieh, Pei-Xuan
Kang, Yihuang
Multiagent Systems
Artificial Intelligence
Computers and Society
Recent advancements in Large Language Models offer promising capabilities to simulate complex human social interactions. We investigate whether LLM-based multi-agent simulations can reproduce core human social dynamics observed in online forums. We evaluate conformity dynamics, group polarization, and fragmentation across different model scales and reasoning capabilities using a structured simulation framework. Our findings indicate that smaller models exhibit higher conformity rates, whereas models optimized for reasoning are more resistant to social influence.
title Towards Simulating Social Influence Dynamics with LLM-based Multi-agents
topic Multiagent Systems
Artificial Intelligence
Computers and Society
url https://arxiv.org/abs/2507.22467