基于状态接地动态检索的Web代理在线技能学习
Online Skill Learning for Web Agents via State-Grounded Dynamic Retrieval
AI总结：
提出状态接地动态检索（SGDR）方法，通过逐步技能重用提升Web代理在多步自动化任务中的表现，在WebArena上平均成功率分别达到37.5%（GPT-4.1）和24.3%（Qwen3-4B）。
摘要：
语言代理越来越依赖可重用技能来改进跨相关任务的多步Web自动化。越来越多的研究关注在线技能学习，其中代理不断从先前的任务轨迹中归纳技能，并在未来的任务中动态重用它们。然而，现有方法主要在任务级别重用技能：根据初始任务指令检索一组固定的技能，并在执行过程中保持不变。这种静态策略与Web执行不一致，因为适当的下一步动作不仅取决于任务目标，还取决于当前网页状态，而网页状态通常会转变为初始技能无法覆盖的情况。为了解决这一差距，我们提出了状态接地动态检索（SGDR），一种在线技能学习方法，使Web代理能够逐步重用技能。SGDR由三个组件组成：一个滑动窗口提取过程，将完成的轨迹转化为可在中间执行状态调用的可重用子程序；一种双文本代码表示，将技能检索与可执行动作连接起来；以及一种状态接地动态检索机制，将技能与任务目标和当前网页状态相匹配。在WebArena上跨五个领域的实验表明，SGDR consistently outperforms strong baselines, achieving average success rates of 37.5% with GPT-4.1 and 24.3% with Qwen3-4B, corresponding to relative gains of 10.6% and 10.0% over the strongest baseline, respectively. 代码可在 https://github.com/plusnli/skill-dynamic-retrieval 获取。
英文摘要：
Language agents increasingly rely on reusable skills to improve multi-step web automation across related tasks. A growing line of work studies online skill learning, where agents continually induce skills from previous task trajectories and reuse them in future tasks on the fly. However, existing methods mainly reuse skills at the task-level: a fixed set of skills is retrieved based on the initial task instruction and then held fixed throughout execution. This static strategy is misaligned with web execution, where the appropriate next action depends not only on the task goal but also on the current webpage state, which often transitions into situations that the initial skills fail to cover. To address this gap, we propose State-Grounded Dynamic Retrieval (SGDR), an online skill learning method that enables stepwise skill reuse for web agents. SGDR consists of three components: a sliding-window extraction process that turns completed trajectories into reusable sub-procedures invokable at intermediate execution states, a dual text-code representation that connects skill retrieval with executable action, and a state-grounded dynamic retrieval mechanism that matches skills to both the task goal and the current webpage state. Experiments on WebArena across five domains show that SGDR consistently outperforms strong baselines, achieving average success rates of 37.5% with GPT-4.1 and 24.3% with Qwen3-4B, corresponding to relative gains of 10.6% and 10.0% over the strongest baseline, respectively. The code is available at https://github.com/plusnli/skill-dynamic-retrieval.
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。
