摘要：大语言模型（LLM）正日益被部署于多智能体环境中。然而，信念如何在相互交互的 LLM 之间形成和传播，仍缺乏充分理解。我们提出 CoevolveSim，这是一个用于研究网络化 LLM 群体中信念扩散的框架。CoevolveSim 使我们能够分离并研究三个因素：领域专长、社会角色分配以及社会网络结构。在该框架内，通用型和专用型 LLM 智能体会交换并修正信念。在每一轮中，LLM 智能体会先观察其邻居信念的摘要，再更新自身信念。我们开展了 1,280 次受控模拟，覆盖 4 种场景、2 类网络结构和 20 条医疗适应症陈述。我们发现，Persona 风格的角色分配和网络结构会重塑个体信念修正，但对群体层面的共识影响极小。相比之下，引入经过微调的专用型 LLM，可使共识变化幅度增加一倍以上，并导致持续存在的影响力施加不对称现象。我们进一步表明，基于简单持久性的观点动力学模型，可以复现全通用型 LLM 群体中的集体结果；而对于异质 LLM 群体，要复现共识需要纳入群体层面的信念构成，要预测个体信念转变则需要考虑智能体身份。我们的结果表明，若要真实模拟多智能体 LLM 系统中的信念扩散，需要采用多样化的底层 LLM，而非仅使用 Persona 提示。

Abstract: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.