Surfer-H 结合 Holo1：基于开放权重的低成本网络代理
人工智能
2025-06-12 v2
摘要
我们提出 Surfer-H，一个低成本的网络代理，集成视觉语言模型（VLM）以在网络上执行用户定义的任务。我们将其与 Holo1 配对，Holo1 是一个新的开放权重 VLM 集合，专门用于网页导航和信息提取。Holo1 在精心策划的数据源上进行训练，包括开放获取的网页内容、合成示例和自生成的代理数据。Holo1 在通用用户界面（UI）基准测试以及我们新的网页 UI 定位基准测试 WebClick 上位居榜首。在 Holo1 的驱动下，Surfer-H 在 WebVoyager 上实现了 92.2% 的最先进性能，在准确性和成本效率之间取得了帕累托最优的平衡。为加速代理系统的研究进展，我们开源了我们的 WebClick 评估数据集和 Holo1 模型权重。
引用
@article{arxiv.2506.02865,
title = {Surfer-H Meets Holo1: Cost-Efficient Web Agent Powered by Open Weights},
author = {Mathieu Andreux and Breno Baldas Skuk and Hamza Benchekroun and Emilien Biré and Antoine Bonnet and Riaz Bordie and Nathan Bout and Matthias Brunel and Pierre-Louis Cedoz and Antoine Chassang and Mickaël Chen and Alexandra D. Constantinou and Antoine d'Andigné and Hubert de La Jonquière and Aurélien Delfosse and Ludovic Denoyer and Alexis Deprez and Augustin Derupti and Michael Eickenberg and Mathïs Federico and Charles Kantor and Xavier Koegler and Yann Labbé and Matthew C. H. Lee and Erwan Le Jumeau de Kergaradec and Amir Mahla and Avshalom Manevich and Adrien Maret and Charles Masson and Rafaël Maurin and Arturo Mena and Philippe Modard and Axel Moyal and Axel Nguyen Kerbel and Julien Revelle and Mats L. Richter and María Santos and Laurent Sifre and Maxime Theillard and Marc Thibault and Louis Thiry and Léo Tronchon and Nicolas Usunier and Tony Wu},
journal= {arXiv preprint arXiv:2506.02865},
year = {2025}
}
备注
Alphabetical order