对比单/多智能体系统，发现大模型进步后混合用更高效
Single-agent or Multi-agent Systems? Why Not Both?
- 提出请求级混合架构，在单/多智能体间动态切换
- 大模型能力提升后，多智能体优势减弱，性能差距缩小
- 兼顾效率与准确率，部署成本降20%，准确率提1.1-12%
多智能体系统（MAS）通过分解任务、分配子任务给不同大型语言模型（LLM）智能体和工具来提升复杂任务处理能力。已有研究显示，MAS在多个领域表现更优，得益于长时上下文追踪和角色专用的错误修正机制。然而，相比单智能体系统（SAS），MAS设计与部署复杂度更高、运行成本更大。与此同时，前沿大模型如OpenAI-o3和Gemini-2.5-Pro在长上下文推理、记忆保留和工具使用方面迅速进步，缓解了早期促使MAS出现的诸多局限。本文对多种主流智能体应用进行了广泛实证比较，发现随着大模型能力提升，MAS相较于SAS的优势逐渐减弱。我们提出了高效机制以定位多智能体系统中的易错环节，并据此设计了一种混合智能体范式——请求级级联（request cascading），在多智能体与单智能体间实现动态切换。该设计在各类智能体应用中提升了1.1%-12%的准确率，同时部署成本降低最高达20%。
原文摘要 · Abstract (English)
Multi-agent systems (MAS) decompose complex tasks and delegate subtasks to different large language model (LLM) agents and tools. Prior studies have reported the superior accuracy performance of MAS across diverse domains, enabled by long-horizon context tracking and error correction through role-specific agents. However, the design and deployment of MAS incur higher complexity and runtime cost compared to single-agent systems (SAS). Meanwhile, frontier LLMs, such as OpenAI-o3 and Gemini-2.5-Pro, have rapidly advanced in long-context reasoning, memory retention, and tool usage, mitigating many limitations that originally motivated MAS designs. In this paper, we conduct an extensive empirical study comparing MAS and SAS across various popular agentic applications. We find that the benefits of MAS over SAS diminish as LLM capabilities improve, and we propose efficient mechanisms to pinpoint the error-prone agent in MAS. Furthermore, the performance discrepancy between MAS and SAS motivates our design of a hybrid agentic paradigm, request cascading between MAS and SAS, to improve both efficiency and capability. Our design improves accuracy by 1.1-12% while reducing deployment costs by up to 20% across various agentic applications.
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。