Improve dataset card: Add paper/code links, update size category, add task/tags

This PR improves the dataset card by:
- Linking to the associated paper: [MST-Distill: Mixture of Specialized Teachers for Cross-Modal Knowledge Distillation](https://huggingface.co/papers/2507.07015)
- Linking to the official code repository: https://github.com/gray1y/MST-Distill
- Correcting the `size_categories` to `10K<n<100K` to accurately reflect the number of samples (48,755).
- Adding the `task_categories: other` and relevant `tags` (`cross-modal`, `knowledge-distillation`, `audio-visual`, `multimodal`, `vggsound`, `feature-extraction`) for better discoverability and context.

README.md

@@ -1,16 +1,27 @@
 ---
 license: cc-by-4.0
 size_categories:
-- 10B<n<100B
+- 10K<n<100K
+task_categories:
+- other
+tags:
+- cross-modal
+- knowledge-distillation
+- audio-visual
+- multimodal
+- vggsound
+- feature-extraction
 ---
 
 # VGGSound-50k Preprocessed Dataset
 
+[Paper](https://huggingface.co/papers/2507.07015) | [Code](https://github.com/gray1y/MST-Distill)
+
 This dataset contains preprocessed data from the VGGSound dataset, specifically processed using the VGGSound-AVEL50k subset for cross-modal knowledge distillation research. The preprocessing is optimized for MST-Distill (Mixture of Specialized Teachers for Cross-Modal Knowledge Distillation) method.
 
 This preprocessing work is based on the VGGSound-AVEL50k subset from: **jasongief/CPSP: [2023 TPAMI] Contrastive Positive Sample Propagation along the Audio-Visual Event Line**
 
-And related preprocessing works are described in our paper: https://arxiv.org/abs/2507.07015
+And related preprocessing works are described in our paper: [MST-Distill: Mixture of Specialized Teachers for Cross-Modal Knowledge Distillation](https://huggingface.co/papers/2507.07015)
 
 ---