Obálka

AgentTrek: Agent Trajectory Synthesis via Guiding Replay with Web Tutorials

Uloženo v:
Vydáno v:arXiv.org (Dec 12, 2024), p. n/a
Hlavní autor: Xu, Yiheng
Další autoři: Lu, Dunjie, Shen, Zhennan, Wang, Junli, Wang, Zekun, Mao, Yuchen, Xiong, Caiming, Yu, Tao
Vydáno:
Cornell University Library, arXiv.org
Témata:
User interface
Visual tasks
Graphical user interface
Annotations
Applications programs
Synthesis
Online tutorials
Trajectory planning
Task complexity
Training
On-line přístup:Citation/Abstract
Full text outside of ProQuest
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstrakt:Graphical User Interface (GUI) agents hold great potential for automating complex tasks across diverse digital environments, from web applications to desktop software. However, the development of such agents is hindered by the lack of high-quality, multi-step trajectory data required for effective training. Existing approaches rely on expensive and labor-intensive human annotation, making them unsustainable at scale. To address this challenge, we propose AgentTrek, a scalable data synthesis pipeline that generates high-quality GUI agent trajectories by leveraging web tutorials. Our method automatically gathers tutorial-like texts from the internet, transforms them into task goals with step-by-step instructions, and employs a visual-language model agent to simulate their execution in a real digital environment. A VLM-based evaluator ensures the correctness of the generated trajectories. We demonstrate that training GUI agents with these synthesized trajectories significantly improves their grounding and planning performance over the current models. Moreover, our approach is more cost-efficient compared to traditional human annotation methods. This work underscores the potential of guided replay with web tutorials as a viable strategy for large-scale GUI agent training, paving the way for more capable and autonomous digital agents.
ISSN:2331-8422
Zdroj:Engineering Database