Large Language Models (LLMs) have demonstrated an unprecedented ability to simulate human-like social behaviors, making them useful tools for simulating complex social systems. However, it remains unclear to what extent these simulations can be trusted to accurately capture key social mechanisms, particularly in highly unbalanced contexts involving minority groups. This paper uses a network generation model with controlled homophily and class sizes to examine how LLM agents behave collectively in multi-round debates. Moreover, our findings highlight a particular directional susceptibility that we term \textit{agreement drift}, in which agents are more likely to shift toward specific positions on the opinion scale. Overall, our findings highlight the need to disentangle structural effects from model biases before treating LLM populations as behavioral proxies for human groups.

翻译：大型语言模型（LLM）已展现出模拟人类社交行为的空前能力，使其成为模拟复杂社会系统的有效工具。然而，这些模拟在多大程度上能精准捕捉关键社会机制（尤其是在涉及少数群体的高度不平衡情境中）仍不明确。本文采用具备可控同质性及群体规模特征的网络生成模型，系统考察LLM主体在多轮辩论中的群体行为模式。此外，研究发现存在特定方向敏感性，我们将其定义为\textit{立场偏移}现象——即主体更倾向沿意见量尺向特定位置迁移。总体而言，本研究揭示出：在将LLM群体作为人类群体的行为代理之前，必须解构结构效应与模型偏见的交互影响。