UNIBROWSE: A Data-to-Agent Framework for Multimodal BrowseComp

Jul 14, 2026

UNIBROWSE introduces a unified data pipeline that, for the first time, generates training data covering all three key information-flow patterns in multimodal browsing: text-only, image-to-text, and text-to-image. The framework augments knowledge graphs with live web retrieval and uses a novel exploration degree metric to filter low-signal data, resulting in high-quality training instances. The trained 35B-parameter agent achieves state-of-the-art performance on multimodal BrowseComp benchmarks, with an average accuracy of 54.4, outperforming several closed-source models including GPT-5 and Gemini-2.5.

Why it matters: This work fills a major gap in multimodal browsing by enabling agents to handle the previously neglected text-to-image pattern, advancing the generality and robustness of web agents.

Full story at: arXiv Computation and Language ↗