medical VQA datasets

The 'medical VQA datasets' consist of visual question answering benchmarks specifically designed for the medical domain, used to evaluate the performance and robustness of vision-language models in answering medical-related questions based on visual inputs.

Papers using medical VQA datasets (4)

MedVR: Annotation-Free Medical Visual Reasoning via Agentic Reinforcement Learning2026ClinCoT: Clinical-Aware Visual Chain-of-Thought for Medical Vision Language Models2026To Agree or To Be Right? The Grounding-Sycophancy Tradeoff in Medical Vision-Language Models2026Self-supervised vision-language pretraining for Medical visual question answering2022 · 10 cites