Problem
Research questions and friction points this paper is trying to address.
browser agents
real-world deployment
long-horizon tasks
complex UIs
alignment
Wuying-Browser-Agent: Real-World Centric Fundamental Long-Horizon Browser Agents
🤖 AI Summary
本文提出Wuying-Browser-Agent框架，通过执行、监督、优化和评估等多层面改进，解决浏览器代理在复杂真实环境中的长周期决策问题。
📝 Abstract
Browser agents perform well on short, clean demonstrations, but real deployment is fundamentally different: agents must sustain dozens of decisions on live websites while recovering from mistakes and navigating complex UIs. We argue that closing this gap requires alignment at every level of the pipeline, including execution, supervision, optimization, and evaluation, rather than scale alone. We present Wuying-Browser-Agent, a unified framework that addresses each of these levels. A structured browser harness provides stable execution primitives and decision-oriented context management. Reflection and UI-specialized Curriculum SFT (RUIC-SFT) explicitly trains on recovery trajectories and complex-UI interactions. Divergence-Aware Online GRPO (DAO-GRPO) improves long-horizon credit assignment through potential-based reward shaping and divergence-aware step weighting. Finally, we introduce BrowserBench, a bilingual real-web benchmark of 350 tasks averaging 37.9 steps, because most existing benchmarks are too short to expose long-horizon failure modes. Wuying-Browser-Agent-27B achieves 80.6\% on WebVoyager, 66.7\% on Online-Mind2Web, and 65.1\% on BrowserBench, establishing a new open-source state of the art on browser-use benchmarks. The same pipeline also transfers beyond browser use, demonstrating strong general agentic ability and reaching an average score of 73.8 on Tau2-Bench, Claw-Eval, and BFCL-v4.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Wuying-Browser-Agent
RUIC-SFT
DAO-GRPO
BrowserBench
🔎 Similar Papers
No similar papers found.
