PMC-VQA is a large-scale medical visual question-answering dataset that contains 227k VQA pairs of 149k images that cover various modalities or diseases. The question-answer pairs are generated from PMC-OA.
Variants: PMC-VQA
This dataset is used in 2 benchmarks:
| Task | Model | Paper | Date |
|---|---|---|---|
| Visual Question Answering (VQA) | MedVInT | PMC-VQA: Visual Instruction Tuning for … | 2023-05-17 |
| Generative Visual Question Answering | MedVInT | PMC-VQA: Visual Instruction Tuning for … | 2023-05-17 |
| Visual Question Answering (VQA) | PMC-CLIP | PMC-CLIP: Contrastive Language-Image Pre-training using … | 2023-03-13 |
| Visual Question Answering (VQA) | BLIP-2 | BLIP-2: Bootstrapping Language-Image Pre-training with … | 2023-01-30 |
| Generative Visual Question Answering | BLIP-2 | BLIP-2: Bootstrapping Language-Image Pre-training with … | 2023-01-30 |
| Generative Visual Question Answering | Open-Flamingo | Flamingo: a Visual Language Model … | 2022-04-29 |
| Visual Question Answering (VQA) | Open-Flamingo | Flamingo: a Visual Language Model … | 2022-04-29 |
Recent papers with results on this dataset: