Publication Type

Conference Proceeding Article

Version

publishedVersion

Publication Date

8-2025

Abstract

Automating web navigation which aims to build a web agent that follows user instructions to complete tasks like booking flights by interacting with websites, has received increasing attention due to its practical value. Although existing web agents are mostly equipped with visual perception, planning, and memory abilities, their reasoning process are still deviate from human cognition. In this work, we study the human thought pattern to empower agent with more human-like abilities in web navigation. To tackle this problem, we propose a novel multimodal web agent framework called WebExperT, which is designed to emulate the human planning process of “thinking fast and slow” to effectively decompose complex user instructions. Furthermore, WebExperT leverages experiential learning by reflecting from failure for continuously refining planning and decision-making outcomes. Experimental results on the Mind2Web benchmark demonstrate the superiority of WebExperT in both supervised and unsupervised settings.

Discipline

Artificial Intelligence and Robotics | Programming Languages and Compilers

Research Areas

Intelligent Systems and Optimization

Areas of Excellence

Digital transformation

Publication

Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics, Vienna, Austria, 2025 July 27 - August 1

First Page

14232

Last Page

14251

Identifier

10.18653/v1/2025.acl-long.697

Publisher

Association for Computational Linguistics

City or Country

USA

Additional URL

https://doi.org/10.18653/v1/2025.acl-long.697

Download

Share

COinS