发表机构
Nanyang Technological University; University of Electronic Science and Technology of China（南洋理工大学; 电子科技大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究通过世界价值观调查和置换检验发现，大型语言模型利用人口统计画像进行文化适应时，会以牺牲个体独特性为代价，将回答拉向群体中心，即“通过刻板印象对齐”，并验证了分散人口统计信号可部分缓解此问题。
AI 中文摘要
大型语言模型越来越多地被部署用于个性化交互，而通过用户画像进行的人口统计条件化是文化适应中广泛采用的策略。我们探究这种方法是否真正服务于个体用户，还是通过抹除个体独特性来实现准确性。我们研究了包括前沿的GPT-5.1在内的七个模型在世界价值观调查上的表现，发现人口统计画像能提高大多数模型的价值对齐准确性，但以系统性的个体性代价为代价。也就是说，模型将回答拉向人口统计群体中心，而非保留个体差异，我们将这种行为模式称为“通过刻板印象对齐”。置换检验（10,000次置换，六个人口统计属性，七个模型）证明，表现最好的模型对个体的压缩程度远高于人类基线；家族内规模扩展加剧了这一权衡，同时降低了内在的文化理解能力。利用一个在来自PRISM（Kirk等人，2024）的真实人机对话上验证的合成对话数据集，我们进一步表明，与紧凑的人口统计标签相比，将人口统计信号分散到对话轮次中能部分抑制原型检索，这一发现在通过PRISM验证的真实对话上得到证实，但需要在更大规模上重复验证。
英文摘要
Large language models are increasingly deployed for personalized interaction, and demographic conditioning via user profiles is a widely adopted strategy for cultural adaptation. We ask whether this approach genuinely serves individual users or achieves accuracy by erasing individual distinctiveness. Studying seven models including frontier GPT-5.1 on the World Values Survey, we find that demographic profiles improve value alignment accuracy for most models, but at a systematic cost to individuality. That is, models pull responses toward demographic group centroids rather than preserving individual differences, a behavioral pattern we term alignment by stereotyping. Permutation tests (10,000 permutations, six demographic attributes, seven models) certify that top-performing models compress individuals far above the human baseline; within-family scaling amplifies this tradeoff while degrading intrinsic cultural understanding. Using a synthetic dialogue dataset validated on real human-chatbot conversations from PRISM (Kirk et al., 2024), we further show that distributing demographic signals across conversational turns partially suppresses prototype retrieval compared to compact demographic labels, a finding validated on real conversations via PRISM but requiring replication at larger scale.
CommentsEMNLP 2026 Findings
