VisualTreeSearch: Understanding Web Agent Test-Time Scaling
Abstract
We present VisualTreeSearch, a fully-deployed system for visualizing and understanding web agent test-time scaling. While test-time search algorithms substantially improve web agent success rates, they remain confined to research contexts with limited practical deployment. Our system bridges this gap with three key contributions: (1) a production-ready solution with cloud-based architecture, (2) an efficient API-based state reset mechanism that reduces state reset time from 50 to 2 s, and (3) an interactive web UI that transparently demonstrates the agent’s decision-making process. VisualTreeSearch provides an intuitive framework for both researchers and users to understand tree search execution in web agents.
Cite
Text
Zhang et al. "VisualTreeSearch: Understanding Web Agent Test-Time Scaling." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2025. doi:10.1007/978-3-032-06129-4_38Markdown
[Zhang et al. "VisualTreeSearch: Understanding Web Agent Test-Time Scaling." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2025.](https://mlanthology.org/ecmlpkdd/2025/zhang2025ecmlpkdd-visualtreesearch/) doi:10.1007/978-3-032-06129-4_38BibTeX
@inproceedings{zhang2025ecmlpkdd-visualtreesearch,
title = {{VisualTreeSearch: Understanding Web Agent Test-Time Scaling}},
author = {Zhang, Danqing and Qian, Yaoyao and He, Shiying and Wang, Yuanli and Ni, Jingyi and Cao, Junyu},
booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
year = {2025},
pages = {497-501},
doi = {10.1007/978-3-032-06129-4_38},
url = {https://mlanthology.org/ecmlpkdd/2025/zhang2025ecmlpkdd-visualtreesearch/}
}