构建可自主网页搜索的智能体，提升复杂问题求解能力
WebDancer: Towards Autonomous Information Seeking Agency
- 从数据出发设计四阶段训练流程，支持端到端智能体开发
- 在GAIA和WebWalkerQA上表现优异，显著优于基线方法
- 适合对自主智能体、多步推理感兴趣的开发者与研究者
解决复杂的现实问题需要深入的信息获取和多步推理。近年来以Deep Research为代表的代理系统展示了自主多步研究的潜力。本文提出一种以数据为中心、面向训练阶段的完整范式，用于构建端到端的代理型信息获取智能体。该方法包含四个关键阶段：(1) 浏览数据构建，(2) 轨迹采样，(3) 监督微调实现有效冷启动，(4) 强化学习提升泛化能力。我们在基于ReAct架构的网页代理WebDancer中实现了这一框架。在GAIA和WebWalkerQA两个挑战性信息获取基准上的实证评估表明，WebDancer表现出色，验证了该训练范式的有效性。对代理训练过程的进一步分析提供了有价值的洞见和可操作的路径，助力开发更强大的代理模型。代码与演示将发布于https://github.com/Alibaba-NLP/WebAgent。
原文摘要 · Abstract (English)
Addressing intricate real-world problems necessitates in-depth information seeking and multi-step reasoning. Recent progress in agentic systems, exemplified by Deep Research, underscores the potential for autonomous multi-step research. In this work, we present a cohesive paradigm for building end-to-end agentic information seeking agents from a data-centric and training-stage perspective. Our approach consists of four key stages: (1) browsing data construction, (2) trajectories sampling, (3) supervised fine-tuning for effective cold start, and (4) reinforcement learning for enhanced generalisation. We instantiate this framework in a web agent based on the ReAct, WebDancer. Empirical evaluations on the challenging information seeking benchmarks, GAIA and WebWalkerQA, demonstrate the strong performance of WebDancer, achieving considerable results and highlighting the efficacy of our training paradigm. Further analysis of agent training provides valuable insights and actionable, systematic pathways for developing more capable agentic models. The codes and demo will be released in https://github.com/Alibaba-NLP/WebAgent.
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。