Paper 2509.13312
WebWeaver: Structuring Web-Scale Evidence with Dynamic Outlines for Open-Ended Deep Research
- Published
- Sep 2025
- Research lab
- Independent
- Citations
- 39
- GitHub
- Not linked
01 In brief
Summary
WebWeaver is a dual-agent framework for open-ended deep research (OEDR), addressing limitations of static pipelines and monolithic generation.
It comprises a planner that iteratively interleaves evidence acquisition with outline optimization, producing a citation-grounded outline linked to a memory bank, and a writer that performs hierarchical, section-by-section synthesis with targeted retrieval.
The framework achieves state-of-the-art results on DeepResearch Bench, DeepConsult, and DeepResearchGym, with citation accuracy of 93.37% and top scores on other benchmarks.
Ablations show iterative outline optimization and hierarchical writing significantly improve quality.
A curated SFT dataset, WebWeaver-3k, enables smaller models to reach expert-level performance, improving citation accuracy from 25% to 85.90%.
02 From the paper
Abstract
This paper tackles \textbf{open-ended deep research (OEDR)}, a complex challenge where AI agents must synthesize vast web-scale information into insightful reports. Current approaches are plagued by dual-fold limitations: static research pipelines that decouple planning from evidence acquisition and monolithic generation paradigms that include redundant, irrelevant evidence, suffering from hallucination issues and low citation accuracy. To address these challenges, we introduce \textbf{WebWeaver}, a novel dual-agent framework that emulates the human research process. The planner operates in a dynamic cycle, iteratively interleaving evidence acquisition with outline optimization to produce a comprehensive, citation-grounded outline linking to a memory bank of evidence. The writer then executes a hierarchical retrieval and writing process, composing the report section by section. By performing targeted retrieval of only the necessary evidence from the memory bank via citations for each part, it effectively mitigates long-context issues and citation hallucinations. Our framework establishes a new state-of-the-art across major OEDR benchmarks, including DeepResearch Bench, DeepConsult, and DeepResearchGym. These results validate our human-centric, iterative methodology, demonstrating that adaptive planning and focused synthesis are crucial for producing comprehensive, trusted, and well-structured reports.