发表机构
Korea Advanced Institute of Science and Technology (KAIST); Seoul National University（韩国科学技术院; 首尔大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究用韩国人格智能体模拟审议，发现其难以再现人口意见模式，且立场变化多非源于互动，表明人口表征与互动驱动变化难以兼得。
AI 中文摘要
多智能体LLM审议已被探索为一种可扩展的模拟公众审议的方式。要使此类模拟具有信息量，人格智能体应反映人口意见模式，且互动应塑造其结论。我们使用基于人口普查的韩国人格智能体就真实政策问题进行辩论，并以全国调查为基准，评估基于LLM的审议能否满足这两个条件。人格智能体不能可靠地再现人口意见模式：其回应往往过于集中，且经常逆转人类数据中的人口统计学差异。尽管如此，审议仍能产生有推理、相互回应且多样的论点，同时伴随显著的立场转变。然而，这种转变中的很大一部分并不需要同伴交流：密封独白智能体以相似速率改变立场，并达到与完整辩论几乎相同的最终平衡，而初始立场迥异的群体往往收敛到相似的终点。与此同时，锚定人口信息起始位置会急剧抑制更新。因此，人口表征、论点生成和互动驱动的意见变化并不必然同时发生。模拟能轻松呈现双方论点，尽管它们是否捕捉了人类观点的多样性仍未得到检验，这为论点呈现留下了一个有前景的角色，即使人口模拟仍需进一步验证。
英文摘要
Multi-agent LLM deliberation has been explored as a scalable way to simulate public deliberation. For such simulations to be informative, persona agents should reflect population opinion patterns and interaction should shape their conclusions. We evaluate whether LLM-based deliberation can meet these two conditions using census-grounded Korean personas debating real policy questions benchmarked against national surveys. Persona agents do not reliably reproduce population opinion patterns: responses are often far more concentrated and frequently reverse demographic differences in the human data. Deliberations nonetheless produce reasoned, reciprocal, and varied arguments alongside substantial stance movement. Yet much of this movement does not require peer exchange: sealed-monologue agents change position at similar rates and reach nearly the same final balance as full debates, while groups initialized with very different positions often converge to similar endpoints. Anchoring population-informed starting positions, meanwhile, sharply suppresses updating. Thus, population representation, argument generation, and interaction-driven opinion change do not necessarily go together. The simulations readily surface arguments on both sides, though whether they capture the diversity of human perspectives remains untested, leaving open a promising role for argument surfacing even as population simulation requires further validation.
