一致却有偏：多智能体临床决策中主席人格以偏差换噪声
原标题：Consistent but Biased: Chair Personas Trade Noise for Bias in Multi-Agent Clinical Decisions
AI 摘要
该研究构建了一个由七个LLM智能体组成的模拟脊柱外科科室，让它们讨论20个退行性颈椎脊髓病病例，病例证据在每次运行中保持不变，仅改变主席及副主席的人格提示（无、协商型、专制型、回避型，分别以行为规则或性格特质形式书写）。结果显示，以规则形式书写时，专制型主席在93%的运行中选择立即手术，回避型仅2%，无人格时为57%，但几乎所有决策仍计划手术，人格主要改变的是立即手术还是先做更多检查；专制和回避人格提高了三次运行间的一致性，但仅与随机水平相当，Fleiss系数未上升。作者建议研究应像变化模型一样报告并变化主席人格，并在称运行间一致性为可靠性前进行随机校正。
正文节选
Consistent but Biased: Chair Personas Trade Noise for Bias in Multi-Agent Clinical Decisions Abstract In clinical systems built from several large language model (LLM) agents, one agent often chairs the discussion and makes the final decision. Whether the chair’s persona prompt changes that decision, or the agreement between repeated runs that some systems treat as reliability, has not been tested with the case held constant. We ask whether the persona changes the decision or only its timing, ho