PRISM：基于特权兄弟图引导的多智能体拓扑设计
原标题：PRISM: PRIvileged Sibling-guided Multi-Agent Topology Design
AI 摘要
该论文提出 PRISM 方法，用于改进基于大语言模型的多智能体系统（MAS）的协作拓扑生成。现有方法通过对已验证图语料进行持续监督微调（SFT）来训练拓扑生成器，但语料只列出成功图而不排序，导致生成器偏向高频拓扑，且存在训练与推理的轨迹不匹配问题。PRISM 通过特权信息学习：教师模型借助零初始化低秩分支读取同一查询下另一个成功图（部署时不可用），让学生在自身采样的轨迹上匹配教师的逐决策目标。在 GSM8K、MultiArith、SVAMP、AQuA、MMLU、HumanEval 六个基准上，PRISM 优于所有评估的拓扑设计基线。
正文节选
PRISM: PRIvileged Sibling-guided Multi-Agent Topology Design Abstract Multi-agent systems (MAS) based on large language models (LLMs) have become a powerful way to address complex problems, and their effectiveness depends critically on the collaboration topology that decides which agents participate and how they communicate. Recent work learns to generate this topology from the task query, fitting the generator to corpora of execution-verified graphs. The post-training method for such generator