VisBrowse-Bench: Benchmarking Visual-Native Search for Multimodal Browsing Agents
This addresses a gap in benchmarking for researchers and developers working on multimodal AI agents, though it is incremental as it builds on existing evaluation frameworks.
The paper tackles the problem of evaluating visual reasoning in multimodal browsing agents by introducing VisBrowse-Bench, a benchmark with 169 VQA instances, and finds that even top models like Claude-4.6-Opus achieve only 47.6% accuracy.
The rapid advancement of Multimodal Large Language Models (MLLMs) has enabled browsing agents to acquire and reason over multimodal information in the real world. But existing benchmarks suffer from two limitations: insufficient evaluation of visual reasoning ability and the neglect of native visual information of web pages in the reasoning chains. To address these challenges, we introduce a new benchmark for visual-native search, VisBrowse-Bench. It contains 169 VQA instances covering multiple domains and evaluates the models' visual reasoning capabilities during the search process through multimodal evidence cross-validation via text-image retrieval and joint reasoning. These data were constructed by human experts using a multi-stage pipeline and underwent rigorous manual verification. We additionally propose an agent workflow that can effectively drive the browsing agent to actively collect and reason over visual information during the search process. We comprehensively evaluated both open-source and closed-source models in this workflow. Experimental results show that even the best-performing model, Claude-4.6-Opus only achieves an accuracy of 47.6%, while the proprietary Deep Research model, o3-deep-research only achieves an accuracy of 41.1%. The code and data can be accessed at: https://github.com/ZhengboZhang/VisBrowse-Bench