发表机构
Universidad Torcuato Di Tella; Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET)（托尔夸托·迪·特利亚大学; 国家科学技术研究委员会（CONICET））
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究将人类审议范式应用于大型语言模型，发现跨模型审议能降低集体误差并提升个体判断，且模型多样性是发挥这一优势的关键。
AI 中文摘要
对众多外行估计的聚合往往优于单个专家的判断，这一现象被称为“群体的智慧”。虽然这通常归因于估计的独立性，但通过审议可以产生更强的效应：对小型审议小组的共识估计取平均值，其效果优于经典的群体智慧，而且个体判断本身在审议后也会变得更加准确。这些改进是否能迁移到大型语言模型之间的相互审议中，目前尚属未知。在此，我们将先前用于人类参与者的三阶段审议范式，改编后应用于来自三个不同家族的大型语言模型，并在四个现实利害关系递增的领域中进行测试：视觉数值估计（研究1）、机器学习论文的同行评审（研究2）、检测人工智能代理隐藏的恶意行为（研究3），以及针对真实预测市场的体育赛事预测（研究4）。在各个领域中，审议在被动聚合独立响应之外进一步降低了集体误差，并且审议后的个体判断保留了这一集体收益。值得注意的是，这一优势需要模型多样性：由单一模型的克隆体组成的群体并未从审议中获益。这些结果确立了机器审议作为一种通用聚合机制的地位，并指出多样性是一个关键要素。
英文摘要
The aggregation of many lay estimates often outperforms individual expert judgment, a phenomenon known as the wisdom of crowds. While this is usually attributed to the independence of estimates, an even stronger effect arises through deliberation: averaging the consensus estimates of small deliberating groups outperforms the classical wisdom of crowds, with individual judgments themselves also becoming more accurate after deliberation. Whether these improvements transfer to large language models deliberating amongst themselves is unknown. Here we adapt a three-stage deliberation paradigm previously used with human participants for use with large language models from three different families, and test it across four domains of increasing real-world stakes: visual numerical estimation (Study 1), peer review of machine-learning papers (Study 2), detection of hidden malicious behavior by an artificial intelligence agent (Study 3), and sports forecasting against a real prediction market (Study 4). Across domains, deliberation reduced collective error beyond passive aggregation of independent responses, and post-deliberation individual judgments retained this collective gain. Notably, the advantage required model diversity: groups composed of clones of a single model did not benefit from deliberating. These results establish machine deliberation as a general-purpose aggregation mechanism, and point to diversity as an active ingredient.
