Synthesizing Agentic Data for Web Agents with Progressive Difficulty Enhancement Mechanisms
Shrey Pandit, Xuan-Phi Nguyen, Yifei Ming, Austin Xu, Jiayu Wang, Caiming Xiong, Shafiq Joty
- 🏛 Institutions
- Salesforce AI Research, University of Wisconsin-Madison
- 📅 Date
- October 15, 2025
- 📑 Publisher
- arXiv
- 💻 Env
- 🔑 Keywords
TLDR
This paper synthesizes training data for deep-research web agents by progressively increasing question difficulty until a baseline agent fails, then using that agent again for validation and filtering. The resulting corpus is aimed at long-horizon online-tool use rather than browser-native GUI interaction, but it is relevant as adjacent training-data work for agent systems.
Related papers