Weblica: 可扩展且可重复的视觉网络代理训练环境
Weblica: Scalable and Reproducible Training Environments for Visual Web Agents
AI总结：
Weblica提出一种可扩展且可重复的视觉网络代理训练框架，通过HTTP级缓存和LLM环境合成技术，实现大规模多样化的网络环境训练，其最佳模型在多个网络导航基准中表现优异。
摘要：
网络环境复杂、开放且不断变化，使视觉网络代理的训练数据扩展变得困难。现有数据收集方法局限于离线轨迹或少量模拟环境，无法捕捉网络多样性。我们提出Weblica（Web Replica）框架，利用HTTP级缓存捕获和回放稳定的视觉状态并保留交互行为，以及基于LLM的环境合成技术。使用该框架，我们扩展了RL训练到数千种多样化环境和任务。我们的最佳模型Weblica-8B在多个网络导航基准中优于类似规模的开源基线模型，使用更少的推理步骤，具有良好的扩展性，并与API模型竞争。
英文摘要：
The web is complex, open-ended, and constantly changing, making it challenging to scale training data for visual web agents. Existing data collection attempts remain limited to offline trajectories for supervised fine-tuning or a handful of simulated environments for RL training, thus failing to capture web diversity. We propose Weblica (Web Replica), a framework for constructing reproducible and scalable web environments. Our framework leverages 1) HTTP-level caching to capture and replay stable visual states while preserving interactive behavior and 2) LLM-based environment synthesis grounded in real-world websites and core web navigation skills. Using this framework, we scale RL training to thousands of diverse environments and tasks. Our best model, Weblica-8B, outperforms open-weight baselines of similar size across multiple web navigation benchmarks while using fewer inference steps, scales favorably with additional test-time compute, and is competitive with API models.
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。
