CoMed：多LLM推理中路由与协作之间的选择性升级
原标题：COMED: The Missing Middle Between Routing and Collaboration in Multi-LLM Inference
AI 摘要
论文提出 CoMed，一种多 LLM 推理中的选择性协作控制器，位于路由与密集协作之间。作者发现跨模型协作具有非单调性：同伴既能挽救单模型无法解决的错误，也可能破坏原本正确的答案。CoMed 通过锚点自一致性、路由边际和轻量同伴探测，决定接受、验证或升级协作，在 16 个开源权重设置中均优于固定或路由锚点，MedQA 最高提升 10.7 个百分点，并在 HLE 上将 GPT-5.5 从 23.1% 提升至 28.1%。
正文节选
COMED: The Missing Middle Between Routing and Collaboration in Multi-LLM Inference Abstract No single Large Language Model (LLM) is uniformly reliable across queries, motivating multi-model inference systems that either route among models or combine their outputs. However, routing stops after selecting an initial model, while dense collaboration invokes peers on every query. We show that collaboration is non-monotonic: peers can recover failures that no model solves alone, but can also corrupt