LLM代理难以真实模拟人类价值观，初始表现就常出错。
The Failure Happens Before the Drift: The Social Dynamics of Values in LLM Agent Societies
- 用世界价值观调查框架构建跨文化对话模拟，测试代理价值忠实度。
- 超50%代理从一开始就不符合设定的价值观，2-7%在对话中发生漂移。
- 适合研究人机交互、社会模拟的学者关注其局限性与设计改进。
大型语言模型（LLM）代理正被用于社会科学中的模拟研究，但其能否准确反映多样且冲突的人类价值观仍不明确。本文基于世界价值观调查（WVS）构建仿真框架，让具有不同沟通风格的文化多样化代理参与长达数千次、涉及15个议题的对话。在约4,000次对话、1,200个角色及三种模型（GPT-4o、Gemini-2.5-Flash、Gemma-4-E4B）的测试中，评估了价值忠实度、价值漂移和对话真实性。结果显示，超过50%的角色从一开始就未能表达其分配的WVS特征，2-7%在重复对话后出现漂移。去除人口统计信息的消融实验虽提升部分模型的忠实度，但整体趋势未变：模拟值分布仍系统性偏离真实分配。相比人类对话，模拟对话在风格一致性与语义多样性之间呈现不同权衡，常产生内容多样但风格重复的交流。结论表明，当前LLM代理虽可生成合理对话，但在长期保持多元人类价值观方面仍存在根本局限。
原文摘要 · Abstract (English)
Large Language Model (LLM)-based agents are increasingly used as proxies for human participants in social science research, yet it remains unclear whether they can faithfully simulate diverse and conflicting human value systems. We present a World Values Survey (WVS)-grounded simulation framework where culturally diverse agents with different communication styles engage in longitudinal, value-laden discussions. Across approximately 4,000 conversations involving 1,200 personas, 15 topics, and three models (GPT-4o, Gemini-2.5-Flash, and Gemma-4-E4B), we evaluate value faithfulness, value drift, and conversational realism. We find that more than 50\% of personas fail to express their assigned WVS profiles from the outset, while 2-7\% drift after repeated conversations. Ablations removing demographic details improve faithfulness for some models but do not change the broader trend: simulated value distributions still systematically deviate from the assigned WVS profiles. Compared to human discussions, simulated dialogues show a different trade-off between stylistic consistency and semantic diversity, often producing content-wise varied but stylistically repetitive exchanges. These findings suggest that current LLM agents can generate plausible conversations, but remain limited proxies for representing and preserving diverse human value profiles over time.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。