发表机构
Khoury College of Computer Sciences, Northeastern University; Network Science Institute, Northeastern University; Santa Fe Institute（东北大学库尔计算机科学学院; 东北大学网络科学研究所; 圣塔菲研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究提出CoevolveSim框架，模拟通用型与专家型LLM组成的社交网络中信念扩散，发现专家型LLM可显著提升共识转变幅度，模拟多智能体LLM系统信念扩散需多样化底层LLM。
AI 中文摘要
大语言模型（LLM）越来越多地被部署在多智能体环境中，但交互的LLM之间信念形成与传播的过程仍鲜为人知。我们提出CoevolveSim，一个用于研究网络化LLM群体中信念扩散的框架，该框架可分离并研究三个因素：领域专业化、社会角色分配及社交网络结构。在该框架中，通用型与专家型LLM智能体交换并修正信念，每轮中，LLM智能体观察邻居信念的摘要后更新自身信念。我们开展了1280次受控模拟，涵盖4种场景、2种网络结构及20条医疗适应症陈述。研究发现，角色分配的人格风格与网络结构会重塑个体信念修正，但对群体层面共识的影响极小；相比之下，引入（微调后的）专家型LLM使共识转变幅度翻倍以上，且产生施加影响力的持续不对称性。我们进一步表明，基于简单持久性的意见动态模型可复现全通用型LLM群体的集体结果，而异质LLM群体则需要群体层面的信念构成来复现共识，且需智能体身份来预测个体信念转变。我们的结果表明，要实现多智能体LLM系统中信念扩散的真实模拟，需要多样化的底层LLM，而非仅靠人格提示。
英文摘要
Large language models (LLMs) are increasingly deployed in multi-agent environments. However, the processes by which beliefs form and propagate among interacting LLMs remain poorly understood. We introduce CoevolveSim, a framework for studying belief diffusion within networked LLM populations. CoevolveSim allows us to isolate and study three factors: domain specialization, social-role assignment, and social network structure. Within this framework, generalist and specialist LLM agents exchange and revise beliefs. In each round, an LLM agent observes a summary of its neighbors' beliefs before updating its own. We run 1,280 controlled simulations spanning four scenarios, two network structures, and 20 medical-indication statements. We find that persona-style role assignment and network structure reshape individual belief revision but have minimal effect on population-level consensus. In contrast, introducing (finetuned) specialist LLMs more than doubles the shift in consensus and gives rise to consistent asymmetries in exerted influence. We further show that simple persistence-based opinion-dynamics models reproduce collective outcomes in all-generalist LLM populations, whereas heterogeneous LLM populations require population-level belief composition to reproduce consensus and agent identity to predict individual belief transitions. Our results indicate that realistic simulation of belief diffusion in multi-agent LLM systems requires a diverse set of underlying LLMs, not persona prompting alone.
Comments33 pages (14 pages of main text), 7 figures, 14 tables
