Metraipolis › archive › dc_ca6724fc530e
WebHarbor: Docking Real Websites for Evolving GUI Agent Environments — arXiv:2611.03634
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.03634
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T11:04:26.590Z
- SHA-256 of the archived text
b33ba313ef9e257d8691292d9894f833533118306a04ef8011cc016ad1235415- Stable URL of this page
https://metraipolis.com/documento/dc_ca6724fc530e
How to cite
WebHarbor: Docking Real Websites for Evolving GUI Agent Environments — arXiv:2611.03634 — arXiv (en), https://arxiv.org/abs/2611.03634 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ca6724fc530e (Metraipolis archive dc_ca6724fc530e, sha256 b33ba313ef9e257d8691292d9894f833533118306a04ef8011cc016ad1235415)
BibTeX
@misc{dcca6724fc530e,
title = {WebHarbor: Docking Real Websites for Evolving GUI Agent Environments — arXiv:2611.03634},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.03634},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ca6724fc530e; sha256 b33ba313ef9e257d8691292d9894f833533118306a04ef8011cc016ad1235415},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.03634. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
GUI web agents require realistic and interactive environments that remain stable during both evaluation and training. Live website evaluation suffers from changing content, access restrictions, and limited coverage of authenticated workflows, while existing synthetic environments for training often lack functional depth and visual fidelity. We propose WebHarbor, a suite of 153 executable and dockerized website environments that faithfully reconstruct the visual interfaces, content, and functional workflows of real-world web services under deterministic control. We build WebHarbor through task-driven human-agent collaboration, where environments evolve as new tasks expand their functional scope, coding agents implement the required changes, and human contributors audit visual fidelity, functional correctness, and task validity. The resulting suite comprises 3,518 tasks spanning multi-step and post-authentication workflows, where existing works often fail to cover. Beyond reliable evaluation, WebHarbor provides high-fidelity synthetic web environments for cost-efficient reinforcement learning (RL). We demonstrate that a Qwen 4B-level GUI agent trained with RL in WebHarbor generalizes to unseen tasks and performs competitively with leading open-source GUI web agents across four benchmarks. The code and data are publicly available at this https URL . Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2611.03634 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-03T10:59:45.899Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).