Research map / Reinforcement learning / Visual and representation RL
WebSailor-V2: Bridging the Chasm to Proprietary Agents via Synthetic Data and Scalable Reinforcement Learning
Kuan Li, Zhongwang Zhang, Huifeng Yin, Rui Ye, Yida Zhao, Liwen Zhang, Litu Ou, Ding-Chu Zhang, Xixi Wu, Xinmiao Yu, Jialong Wu, Xinyu Wang and 9 more
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add RL stage after SFT cold start (ablation) | WebSailor-V2-30B-A3B · BrowseComp-EN | pass@1 (%): 24.4 -> 35.3 | Improved |
| add RL stage after SFT cold start (ablation) | WebSailor-V2-30B-A3B · BrowseComp-ZH | pass@1 (%): 28.3 -> 44.1 | Improved |
| add RL stage after SFT cold start (ablation) | WebSailor-V2-30B-A3B · HLE | pass@1 (%): 23.9 -> 30.6 | Improved |
| add RL stage after SFT cold start (ablation) | WebSailor-V2-30B-A3B · GAIA | pass@1 (%): 66.0 -> 74.1 | Improved |
| train RL directly on BrowseComp test-style human data instead of synthetic SailorFog-QA-V2 (ablation) | WebSailor-V2 · BrowseComp | accuracy | Hurt |
| exclude unfinished negative trajectories (e.g. over length limit) from RL loss | GRPO variant · training stability | format collapse | Improved |
| increase context from 32k to 128k and 100 ReAct iterations | WebSailor-V2 · BrowseComp-EN | accuracy: about 16 at 32k context | Improved |
Closest papers by meaning
- WebRL: Training LLM Web Agents via Self-Evolving Online Curriculum Reinforcement Learning (ICLR 2025 · similarity 0.91)
- WebGym: Scaling Training Environments for Long-Horizon Visual Web Agents with Realistic Tasks (CVPR 2026 · similarity 0.90)
- WebFactory: Automated Compression of Foundational Language Intelligence into Grounded Web Agents (ICLR 2026 · similarity 0.90)
- How to Train Your LLM Web Agent: A Statistical Diagnosis (NeurIPS 2025 · similarity 0.88)
- Learning Agentic Policy from Action Guidance (NeurIPS 2026 · similarity 0.88)
- A Subgoal-driven RL Framework for Improving Long-Horizon Web Agents (NeurIPS 2026 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
