arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.15384cs.AIcs.CL
收藏

具有动作修正的世界模型增强型网络代理

World-Model-Augmented Web Agents with Action Correction

Zhejiang University(浙江大学);Tencent AI Lab(腾讯AI实验室)

Zhouzhou Shen, Xueyu Hu, Xiyun Li, Tianqing Fang, Juncheng Li, Shengyu Zhang

AI总结:

WAC通过多代理协作和风险感知机制,提升网络代理在复杂任务中的鲁棒性和执行效率。

摘要:

基于大语言模型的网络代理在自动化网络任务方面展现出了有前途的能力。然而,当前的网络代理在推断出合理动作方面存在困难,由于预测环境变化的限制,且可能缺乏对执行风险的全面意识,提前执行有风险的动作导致损失并导致任务失败。为了解决这些挑战,我们提出了WAC,一种集成了模型协作、后果模拟和反馈驱动的动作优化的网络代理。为克服单个模型的认知孤立,我们引入了多代理协作过程,使动作模型可以咨询世界模型作为网络环境专家以获得战略指导;动作模型然后将这些建议转化为可执行的动作,利用对环境状态转换动态的先验知识来增强候选动作提案。为实现风险感知的鲁棒任务执行,我们引入了两阶段推导链。一个专门处理环境状态转换的世界模型模拟动作结果,然后由判断模型审查以在必要时触发动作修正反馈。实验显示,WAC在VisualWebArena上实现了1.8%的绝对收益,在Online-Mind2Web上实现了1.3%的收益。

英文摘要:

Web agents based on large language models have demonstrated promising capability in automating web tasks. However, current web agents struggle to reason out sensible actions due to the limitations of predicting environment changes, and might not possess comprehensive awareness of execution risks, prematurely performing risky actions that cause losses and lead to task failure. To address these challenges, we propose WAC, a web agent that integrates model collaboration, consequence simulation, and feedback-driven action refinement. To overcome the cognitive isolation of individual models, we introduce a multi-agent collaboration process that enables an action model to consult a world model as a web-environment expert for strategic guidance; the action model then grounds these suggestions into executable actions, leveraging prior knowledge of environmental state transition dynamics to enhance candidate action proposal. To achieve risk-aware resilient task execution, we introduce a two-stage deduction chain. A world model, specialized in environmental state transitions, simulates action outcomes, which a judge model then scrutinizes to trigger action corrective feedback when necessary. Experiments show that WAC achieves absolute gains of 1.8% on VisualWebArena and 1.3% on Online-Mind2Web.

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。

↑