![[2602.15384] World-Model-Augmented Web Agents with Action Correction](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.15384] World-Model-Augmented Web Agents with Action Correction
Summary
The paper presents WAC, a web agent that enhances task execution by integrating model collaboration, consequence simulation, and action refinement to improve decision-making and reduce risks.
Why It Matters
As web agents increasingly automate tasks, ensuring they can reason effectively and manage risks is crucial. This research addresses the limitations of current models, potentially leading to more reliable AI applications in various domains.
Key Takeaways
- WAC integrates multi-agent collaboration for improved decision-making.
- The two-stage deduction chain enhances risk awareness in task execution.
- Experimental results show significant performance gains over existing models.
Computer Science > Artificial Intelligence arXiv:2602.15384 (cs) [Submitted on 17 Feb 2026] Title:World-Model-Augmented Web Agents with Action Correction Authors:Zhouzhou Shen, Xueyu Hu, Xiyun Li, Tianqing Fang, Juncheng Li, Shengyu Zhang View a PDF of the paper titled World-Model-Augmented Web Agents with Action Correction, by Zhouzhou Shen and 5 other authors View PDF HTML (experimental) Abstract: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 ac...