当证据塑造协作：面向多智能体系统的知识条件拓扑生成
When Evidence Shapes Collaboration: Knowledge-Conditioned Topology Generation for Multi-Agent Systems
浏览论文内容
中文总结 AI 辅助
该研究针对多智能体系统拓扑生成的结构-知识错位问题，提出K-GAT框架，在GPQA数据集上较LLM-Debate基线准确率提升15.7%且计算开销更低。
中文摘要 AI 辅助
多智能体系统（MAS）近期已从静态工作流转向动态生成的协作拓扑。然而，现有拓扑生成方法主要依赖大语言模型的参数化知识，外部搜索或检索仅作为反应工具，而非协作结构的显式决定因素，这导致结构-知识错位，系统在知识密集型任务中出现冗余交互或验证不足。我们提出K-GAT（知识引导智能体拓扑生成器），一种神经符号框架，将协作拓扑设计表述为知识条件结构学习问题，直接将外部证据整合到自回归图生成中。在知识密集型基准上的大量实验表明K-GAT的效率与有效性：尤其在专家级GPQA数据集上，K-GAT的准确率较LLM-Debate基线大幅提升15.7%，同时消耗的计算令牌不到一半。
英文摘要
Multi-Agent Systems (MAS) have recently moved from static workflows toward dynamically generated collaboration topologies. However, existing topology generation methods rely primarily on the parametric knowledge of large language models, with external search or retrieval used only as a reactive tool rather than an explicit determinant of collaboration structure. This leads to structure-knowledge misalignment, where systems exhibit redundant interactions or insufficient verification in knowledge-intensive tasks. We propose K-GAT (Knowledge-Guided Agent Topology Generator), a neuro-symbolic framework that formulates collaboration topology design as a knowledge-conditioned structure learning problem, integrating external evidence directly into autoregressive graph generation. Extensive experiments on knowledge-intensive benchmarks demonstrate K-GAT's efficiency and effectiveness: notably on the expert-level GPQA dataset, K-GAT outperforms the LLM-Debate baseline by a substantial margin of +15.7% in accuracy, while consuming less than half the computational tokens.
发表机构
- Huazhong University of Science and Technology（华中科技大学）
机构由 AI 辅助整理，请以论文原文为准。
