WebDancer: Towards Autonomous Information Seeking Agency
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912529393909760 |
|---|---|
| author | Wu, Jialong Li, Baixuan Fang, Runnan Yin, Wenbiao Zhang, Liwen Tao, Zhengwei Zhang, Dingchu Xi, Zekun Fu, Gang Jiang, Yong Xie, Pengjun Huang, Fei Zhou, Jingren |
| author_facet | Wu, Jialong Li, Baixuan Fang, Runnan Yin, Wenbiao Zhang, Liwen Tao, Zhengwei Zhang, Dingchu Xi, Zekun Fu, Gang Jiang, Yong Xie, Pengjun Huang, Fei Zhou, Jingren |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_22648 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | WebDancer: Towards Autonomous Information Seeking Agency Wu, Jialong Li, Baixuan Fang, Runnan Yin, Wenbiao Zhang, Liwen Tao, Zhengwei Zhang, Dingchu Xi, Zekun Fu, Gang Jiang, Yong Xie, Pengjun Huang, Fei Zhou, Jingren Computation and Language 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. |
| title | WebDancer: Towards Autonomous Information Seeking Agency |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2505.22648 |