Magentic-One：一个用于解决复杂任务的通用多智能体系统
Magentic-One: A Generalist Multi-Agent System for Solving Complex Tasks
AI总结：
本文提出Magentic-One，一个基于多智能体架构的通用智能体系统，通过编排器协调专门智能体，在GAIA、AssistantBench和WebArena基准上达到最先进性能，并提供开源实现与评估工具AutoGenBench。
摘要：
现代人工智能智能体，受大型基础模型进步的推动，有望通过增强我们的知识和能力来提高我们的生产力并改变我们的生活。为了实现这一愿景，人工智能智能体必须有效地进行规划、执行多步推理和行动、应对新观察，并从错误中恢复，以在广泛场景中成功完成复杂任务。在这项工作中，我们介绍了Magentic-One，一个用于解决此类任务的高性能开源智能体系统。Magentic-One采用多智能体架构，其中主智能体——编排器（Orchestrator）负责规划、跟踪进度并重新规划以从错误中恢复。在整个任务执行过程中，编排器根据需要指导其他专门智能体执行任务，例如操作网页浏览器、导航本地文件或编写和执行Python代码。我们展示了Magentic-One在三个多样且具有挑战性的智能体基准测试：GAIA、AssistantBench和WebArena上，达到了与最先进技术统计上相当的性能。Magentic-One在不修改核心智能体能力或其协作方式的情况下实现了这些结果，展示了向通用智能体系统迈进的进展。此外，Magentic-One的模块化设计允许在不进行额外提示调整或训练的情况下向团队添加或移除智能体，从而简化开发并使其可扩展到未来场景。我们提供了Magentic-One的开源实现，并包含了AutoGenBench，一个用于智能体评估的独立工具。AutoGenBench提供了内置的重复和隔离控制，以严格且受控的方式运行智能体基准测试——这在智能体的行动具有副作用时非常重要。Magentic-One、AutoGenBench以及Magentic-One的详细实证性能评估，包括消融实验和错误分析，可在https://aka.ms/magentic-one获取。
英文摘要：
Modern AI agents, driven by advances in large foundation models, promise to enhance our productivity and transform our lives by augmenting our knowledge and capabilities. To achieve this vision, AI agents must effectively plan, perform multi-step reasoning and actions, respond to novel observations, and recover from errors, to successfully complete complex tasks across a wide range of scenarios. In this work, we introduce Magentic-One, a high-performing open-source agentic system for solving such tasks. Magentic-One uses a multi-agent architecture where a lead agent, the Orchestrator, plans, tracks progress, and re-plans to recover from errors. Throughout task execution, the Orchestrator directs other specialized agents to perform tasks as needed, such as operating a web browser, navigating local files, or writing and executing Python code. We show that Magentic-One achieves statistically competitive performance to the state-of-the-art on three diverse and challenging agentic benchmarks: GAIA, AssistantBench, and WebArena. Magentic-One achieves these results without modification to core agent capabilities or to how they collaborate, demonstrating progress towards generalist agentic systems. Moreover, Magentic-One's modular design allows agents to be added or removed from the team without additional prompt tuning or training, easing development and making it extensible to future scenarios. We provide an open-source implementation of Magentic-One, and we include AutoGenBench, a standalone tool for agentic evaluation. AutoGenBench provides built-in controls for repetition and isolation to run agentic benchmarks in a rigorous and contained manner -- which is important when agents' actions have side-effects. Magentic-One, AutoGenBench and detailed empirical performance evaluations of Magentic-One, including ablations and error analysis are available at https://aka.ms/magentic-one
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
