WebCoT: Enhancing Web Agent Reasoning by Reconstructing Chain-of-Thought in Reflection, Branching, and Rollback

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hu, Minda, Fang, Tianqing, Zhang, Jianshu, Ma, Junyu, Zhang, Zhisong, Zhou, Jingyan, Zhang, Hongming, Mi, Haitao, Yu, Dong, King, Irwin
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911161355599872
author Hu, Minda
Fang, Tianqing
Zhang, Jianshu
Ma, Junyu
Zhang, Zhisong
Zhou, Jingyan
Zhang, Hongming
Mi, Haitao
Yu, Dong
King, Irwin
author_facet Hu, Minda
Fang, Tianqing
Zhang, Jianshu
Ma, Junyu
Zhang, Zhisong
Zhou, Jingyan
Zhang, Hongming
Mi, Haitao
Yu, Dong
King, Irwin
contents Web agents powered by Large Language Models (LLMs) show promise for next-generation AI, but their limited reasoning in uncertain, dynamic web environments hinders robust deployment. In this paper, we identify key reasoning skills essential for effective web agents, i.e., reflection & lookahead, branching, and rollback, and curate trajectory data that exemplifies these abilities by reconstructing the agent's (inference-time) reasoning algorithms into chain-of-thought rationales. We conduct experiments in the agent self-improving benchmark, OpenWebVoyager, and demonstrate that distilling salient reasoning patterns into the backbone LLM via simple fine-tuning can substantially enhance its performance. Our approach yields significant improvements across multiple benchmarks, including WebVoyager, Mind2web-live, and SimpleQA (web search), highlighting the potential of targeted reasoning skill enhancement for web agents.
format Preprint
id arxiv_https___arxiv_org_abs_2505_20013
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle WebCoT: Enhancing Web Agent Reasoning by Reconstructing Chain-of-Thought in Reflection, Branching, and Rollback
Hu, Minda
Fang, Tianqing
Zhang, Jianshu
Ma, Junyu
Zhang, Zhisong
Zhou, Jingyan
Zhang, Hongming
Mi, Haitao
Yu, Dong
King, Irwin
Computation and Language
Web agents powered by Large Language Models (LLMs) show promise for next-generation AI, but their limited reasoning in uncertain, dynamic web environments hinders robust deployment. In this paper, we identify key reasoning skills essential for effective web agents, i.e., reflection & lookahead, branching, and rollback, and curate trajectory data that exemplifies these abilities by reconstructing the agent's (inference-time) reasoning algorithms into chain-of-thought rationales. We conduct experiments in the agent self-improving benchmark, OpenWebVoyager, and demonstrate that distilling salient reasoning patterns into the backbone LLM via simple fine-tuning can substantially enhance its performance. Our approach yields significant improvements across multiple benchmarks, including WebVoyager, Mind2web-live, and SimpleQA (web search), highlighting the potential of targeted reasoning skill enhancement for web agents.
title WebCoT: Enhancing Web Agent Reasoning by Reconstructing Chain-of-Thought in Reflection, Branching, and Rollback
topic Computation and Language
url https://arxiv.org/abs/2505.20013