# Med-R2: An Adversarial Benchmark for Evidence-Grounded Reasoning in Medical VLMs

Source: [Med-R2: An Adversarial Benchmark for Evidence-Grounded Reasoning in Medical VLMs](<https://www.sota2.com/research/paper/6a15292fdbd7150506f7dfc9>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-27T05:00:05.520Z

Authors: Wen Ma, Fucheng Niu, Zhiting Fan, Zikai Xiao, Jiaxiang Liu, Zuozhu Liu

Published: 2026-05-23T09:50:19.000Z

## Abstract

Vision-language models have demonstrated impressive capabilities in general medical visual question answering, yet due to limited interpretability, it remains unclear whether their predictions reflect evidence-grounded clinical reasoning or reliance on spurious priors. We introduce Med-R2 Bench, a hierarchical benchmark aligned with the clinical workflow to evaluate adversarial robustness with visual grounding. We design stepwise QA tasks to assess whether reasoning chains are strictly grounded in visual evidence across the four clinical stages, and employ adversarial perturbations to test robustness against misleading cues. Med-R2 comprises 42,432 images, 31 task categories, and 110,406 QA pairs. Evaluation across 14 VLMs reveals a sequential performance degradation along the four-stage clinical workflow. Adversarial experiments show that models rely heavily on correct prompts to guess answers. Even when provided with explicit visual cues, the models struggle to accurately align textual descriptions. Finally, we demonstrate stepwise fine-tuning using our hierarchical data significantly improves reasoning robustness, highlighting its potential to drive future improvements in evidence-based medical AI.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Medical Report Generation on Med-R2 (test)](<https://www.sota2.com/research/sota/medical-report-generation-on-med-r2-test.md>): Medical Report Generation; Med-R2 (test); ROUGE-1: 14.6.

- [Medical Visual Question Answering on Med-R2 (test)](<https://www.sota2.com/research/sota/medical-visual-question-answering-on-med-r2-test.md>): Medical Visual Question Answering; Med-R2 (test); Image Quality ACC: 47.4.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2605.24492>)

[PDF](<https://arxiv.org/pdf/2605.24492.pdf>)