WebNavigator: Global Web Navigation via Interaction Graph Retrieval

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Zhang, Xuanwang, Han, Yuteng, Qi, Jinnan, Xie, Mulong, Wu, Zhen, Dai, Xinyu
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866915878089523200
author Zhang, Xuanwang
Han, Yuteng
Qi, Jinnan
Xie, Mulong
Wu, Zhen
Dai, Xinyu
author_facet Zhang, Xuanwang
Han, Yuteng
Qi, Jinnan
Xie, Mulong
Wu, Zhen
Dai, Xinyu
contents Despite significant advances in autonomous web navigation, current methods remain far from human-level performance in complex web environments. We argue that this limitation stems from Topological Blindness, where agents are forced to explore via trial-and-error without access to the global topological structure of the environment. To overcome this limitation, we introduce WebNavigator, which reframes web navigation from probabilistic exploration into deterministic retrieval and pathfinding. WebNavigator constructs Interaction Graphs via zero-token cost heuristic exploration offline and implements a Retrieve-Reason-Teleport workflow for global navigation online. WebNavigator achieves state-of-the-art performance on WebArena and OnlineMind2Web. On WebArena multi-site tasks, WebNavigator achieves a 72.9\% success rate, more than doubling the performance of enterprise-level agents. This work reveals that Topological Blindness, rather than model reasoning capabilities alone, is an underestimated bottleneck in autonomous web navigation.
format Preprint
id arxiv_https___arxiv_org_abs_2603_20366
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle WebNavigator: Global Web Navigation via Interaction Graph Retrieval
Zhang, Xuanwang
Han, Yuteng
Qi, Jinnan
Xie, Mulong
Wu, Zhen
Dai, Xinyu
Information Retrieval
Artificial Intelligence
Despite significant advances in autonomous web navigation, current methods remain far from human-level performance in complex web environments. We argue that this limitation stems from Topological Blindness, where agents are forced to explore via trial-and-error without access to the global topological structure of the environment. To overcome this limitation, we introduce WebNavigator, which reframes web navigation from probabilistic exploration into deterministic retrieval and pathfinding. WebNavigator constructs Interaction Graphs via zero-token cost heuristic exploration offline and implements a Retrieve-Reason-Teleport workflow for global navigation online. WebNavigator achieves state-of-the-art performance on WebArena and OnlineMind2Web. On WebArena multi-site tasks, WebNavigator achieves a 72.9\% success rate, more than doubling the performance of enterprise-level agents. This work reveals that Topological Blindness, rather than model reasoning capabilities alone, is an underestimated bottleneck in autonomous web navigation.
title WebNavigator: Global Web Navigation via Interaction Graph Retrieval
topic Information Retrieval
Artificial Intelligence
url https://arxiv.org/abs/2603.20366