| **License:** Apache-2.0 | |
| This dataset is derived from [flaviagiammarino/vqa-rad](https://huggingface.co/datasets/flaviagiammarino/vqa-rad). | |
| All credit goes to the original authors. | |
| The dataset has been reformatted into `.tsv` files for compatibility with [VLMEvalKit](https://github.com/your-vlmevalkit-link). | |
| No changes were made to the original image or text content. | |
| This version is intended for benchmarking vision-language models using VLMEvalKit. | |