发表机构
McGill University; Mila; University of Washington; Apple（麦吉尔大学; Mila; 华盛顿大学; 苹果公司）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文研究多智能体系统中的同质化现象，提出三个度量指标，在代码生成、招聘和同行评审中验证其风险，并发现简单多样性方法无效，需新策略。
AI 中文摘要
多智能体系统（MAS）利用智能体之间的交互来执行复杂任务。尽管取得了成功，但我们表明这些交互也可能导致同质化，即智能体收敛到相似的行为。MAS中的同质化会降低智能体的多样性并强化共同的失败。在本文中，我们使用三个指标来操作化同质化：对多数派的遵从性、向极端的极化以及后续交互中对变化的惯性增长。我们评估了MAS在代码生成、招聘和科学同行评审中的同质化。在这些任务中，我们表明同质化转化为具体的下游风险：在代码生成中，它隐藏并放大了相关错误，这可能造成系统性漏洞；在招聘中，它使得有偏见的智能体的影响在其被移除后仍长期存在；在同行评审中，它造成了不同研究领域之间不均衡的评估标准。我们的结果确立了同质化作为MAS的一种失败模式，表明MAS评估必须超越总体性能，仔细分析交互动态。最后，我们表明增加多样性的简单方法——利用采样随机性和混合模型MAS——未能减少同质化风险，突显了有效利用智能体多样性的策略需求。
英文摘要
Multi-agent systems (MAS) leverage interactions between agents to perform complex tasks. Despite their success, we show that these interactions can also lead to homogenization, i.e., agents converging to similar behaviors. Homogenization in MAS can reduce agent diversity and reinforce shared failures. In this paper, we operationalize homogenization using three metrics: conformity to the majority, polarization towards extremes, and growing inertia against changes over subsequent interactions. We evaluate homogenization in MAS for code generation, hiring, and scientific peer review. Across these tasks, we show that homogenization translates to concrete downstream risks: in code generation, it hides and amplifies correlated errors which can create systemic vulnerabilities; in hiring, it allows the influence of biased agents to persist long after their removal; and in peer review, it creates uneven evaluation standards across research areas. Our results establish homogenization as a failure mode of MAS, demonstrating that MAS evaluations must move beyond aggregate performance to carefully analyze interaction dynamics. Finally, we show that simple approaches to increase diversity---leveraging sampling stochasticity and mixed-models MAS---fail to reduce homogenization risks, highlighting the need for strategies to effectively leverage agent diversity.
