发表机构
Institute for Artificial Intelligence Research and Development of Serbia; University of Belgrade; Institute for Philosophy and Social Theory; Faculty of Political Sciences; University of Vienna; Faculty of Social Sciences; Universitas Negeri Padang; Faculty of Engineering; University of Novi Sad; Faculty of Philosophy; Nanyang Technological University; School of Social Sciences（塞尔维亚人工智能研发研究所; 贝尔格莱德大学; 哲学与社会理论研究所; 政治科学学院; 维也纳大学; 社会科学学院; 巴东国立大学; 工程学院; 诺维萨德大学; 哲学学院; 南洋理工大学; 社会科学学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究通过直觉式提示而非分析式提示，显著提升了LLM智能体模拟个体社交媒体反应的保真度，且在不熟悉内容上同样有效，有望成为通用模拟用户。
AI 中文摘要
平台政策越来越多地在人工用户上测试，这使得智能体的保真度变得重要。然而，令人信服的虚假个人资料也可能在选举前操纵公众认知。验证工作一直集中在与人类行为的一致性上，很少关注智能体的行为是否与其被赋予的个人资料相符。本研究通过问卷、深度访谈和书面自我陈述对八名塞尔维亚参与者进行了画像，记录了他们六十八条社交媒体帖子的反应，并让四个语言模型在五种提示条件下预测这些反应，这些条件在个人资料内容和指令风格上有所不同。态度内容比人口统计学背景信息在预测方面有大幅提升。智能体与其所述个人资料的匹配程度高于参与者与其自身调查回答的匹配程度，且一旦存在个人资料信息，一致性被证明与保真度无关。指示模型直觉地、即时地而非分析性地做出反应，在所有条件下产生了最高的保真度，并将个体差异的压缩从人类水平的七倍降至三倍。这一优势在问卷从未提及的主题的帖子上也得以保持，在该条件下达到了所有设置中的最高保真度，并大幅超过了人群基线，这表明以这种方式提示的智能体可以作为通用模拟用户，而非针对其画像主题的专家。结果可能对语言模型的发展产生影响，因为基于直觉的设置似乎比基于推理的设置更适合某些任务。
英文摘要
Platform policies are increasingly tested on artificial users, making agent fidelity important. Yet convincing fake profiles could also manipulate perceived public opinion before elections. Validation has concentrated on agreement with human behaviour and has paid little attention to whether an agent behaves in line with the profile it was given. The present study profiled eight Serbian participants through a questionnaire, a deep interview, and a written self-presentation, recorded their reactions to sixty-eight social media posts, and asked four language models to predict those reactions under five prompt conditions varying profile content and instruction style. Attitudinal content improved prediction over demographic backstories by a wide margin. Agents matched their stated profiles more closely than participants matched their own survey answers, and consistency proved unrelated to fidelity once profile information was present. Instructing models to respond intuitively and immediately rather than analytically gave the highest fidelity of any condition and cut the compression of individual differences from seven times the human level to three. The advantage held on posts about topics the questionnaire never raised, where that condition reached the highest fidelity of any setup and beat a crowd baseline by a wide margin, which suggests that agents prompted this way could serve as general-purpose simulated users rather than specialists on the topics they were profiled for. Results may bear implications for the development of language models, because intuition-based setups appear better suited to some tasks than reasoning-based ones.
Comments24 pages, 7 figures
