Synthesizing Agentic Data for Web Agents with Progressive Difficulty Enhancement Mechanisms
Abstract
A two-pronged data synthesis pipeline generates complex question-answer pairs, enabling the training of more effective web-based research agents with higher diversity in tool use.
Web-based 'deep research' agents aim to solve complex question - answering tasks through long-horizon interactions with online tools. These tasks remain challenging, as the underlying language models are often not optimized for long-horizon reasoning and exploration. Prior work has proposed workflows for constructing instruction-tuning datasets, often leveraging knowledge graphs. However, such methods typically lack fine-grained control over difficulty and quality, yielding synthetic data that falls short of capturing the complexity required for long-horizon reasoning. Furthermore, many studies conflate data and training effects by comparing models trained under different optimization recipes, making it difficult to isolate and evaluate the effectiveness of the data itself. We introduce a two-pronged data synthesis pipeline that generates question - answer pairs by progressively increasing task complexity until a frontier baseline web agent fails. The baseline agent plays multiple roles in this process: attempting the questions, validating factuality, checking for alternative answers, and enforcing filtering. To evaluate the effectiveness of our synthesis methods, we adopt a controlled training setup based on distillation from strong web agents. Experiments across multiple web-based benchmarks show that our dataset - despite being smaller - enables the training of more effective web agents than existing datasets. In particular, our data exhibits twice the diversity in tool-use actions, allowing models trained on it to achieve stronger performance while avoiding repetitive tool-calling behaviors.
Community
The study presents a controlled data synthesis pipeline for training web-based “deep research” agents that handle complex, multi-step reasoning tasks.
It incrementally generates question–answer pairs until a baseline agent fails, ensuring increasing task complexity and factual accuracy.
The resulting dataset, though smaller, yields more capable agents with greater tool-use diversity and improved performance over prior datasets.
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- DeepDive: Advancing Deep Search Agents with Knowledge Graphs and Multi-Turn RL (2025)
- WebExplorer: Explore and Evolve for Training Long-Horizon Web Agents (2025)
- Open Data Synthesis For Deep Research (2025)
- InfoAgent: Advancing Autonomous Information-Seeking Agents (2025)
- From Evidence to Trajectory: Abductive Reasoning Path Synthesis for Training Retrieval-Augmented Generation Agents (2025)
- Fathom-DeepResearch: Unlocking Long Horizon Information Retrieval and Synthesis for SLMs (2025)
- Explore to Evolve: Scaling Evolved Aggregation Logic via Proactive Online Exploration for Deep Research Agents (2025)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment:
@librarian-bot
recommend
Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper