用可组合组件结构化搜索，让多智能体系统更高效可解释
Grammar Search for Multi-Agent Systems
- 用固定组件构建多智能体系统，替代自由生成代码
- 在五个基准中四胜，效率更高且逻辑更清晰
- 适合追求可解释性和低成本部署的开发者
自动搜索多智能体系统已成为智能体人工智能研究的关键方向。以往方法依赖大模型在代码空间中进行自由生成式搜索。本文提出一种更结构化的框架，通过一组简单、可组合的组件探索相同空间。尽管在候选生成阶段缺乏大模型的生成灵活性，但该方法在数学和问答两个领域的五个基准中的四个上表现优于先前方法。此外，该方法还具备成本更低的搜索过程，并能生成模块化、可解释性强、逻辑更简单的多智能体系统。
原文摘要 · Abstract (English)
Automatic search for Multi-Agent Systems has recently emerged as a key focus in agentic AI research. Several prior approaches have relied on LLM-based free-form search over the code space. In this work, we propose a more structured framework that explores the same space through a fixed set of simple, composable components. We show that, despite lacking the generative flexibility of LLMs during the candidate generation stage, our method outperforms prior approaches on four out of five benchmarks across two domains: mathematics and question answering. Furthermore, our method offers additional advantages, including a more cost-efficient search process and the generation of modular, interpretable multi-agent systems with simpler logic.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。