Add paper link, license

This PR ensures the dataset can be found at https://huggingface.co/papers/2505.21333 and adds a pointer to the license.

README.md

@@ -36,4 +36,11 @@ configs:
   data_files:
   - split: train
     path: data/train-*
+task_categories:
+- video-text-to-text
 ---
+
+This dataset is from the paper [MME-VideoOCR: Evaluating OCR-Based Capabilities of Multimodal LLMs in
+  Video Scenarios](https://huggingface.co/papers/2505.21333).
+
+See https://github.com/DogNeverSleep/MME-VideoOCR_Dataset for more information, including the license.