Update README.md

README.md

@@ -95,6 +95,11 @@ Key results:
 - Vocabulary: **11,424 tags**; ~1M training images; last 100K images held out as validation
 - Noise tags removed via `data/ignored_tags.txt`
 
+## Future plan
+- [ ] Care more about the data (de-duplication, remove noise, etc.). Current data cleaning pipeline is too basic?
+- [ ] Try KD + Transfer Learning from L16 => B16/S16 models.
+- [ ] Another self-attention block for the head => more semantic for the tags.
+
 ## Acknowledgements
 
 - [DINOv3](https://github.com/facebookresearch/dinov3) (Meta AI) — pretrained ViT backbones (licensed under the DINOv3 License)