flyingbagel/mmDocVQA

--- dataset_info: - config_name: default features: - name: id dtype: string - name: media dtype: image - name: messages dtype: string splits: - name: validation num_bytes: 3693856870 num_examples: 5349 - name: test num_bytes: 3781107402 num_examples: 5188 download_size: 11254162628 dataset_size: 7474964272 configs: - config_name: default data_files: - split: test path: data/test-* - split: validation path: data/validation-* --- # mmDocVQA DocVQA in mm-eval format (`id`, `media`, `messages`). See `dataset_info.json` for schema and upstream mapping.