Add pipeline tag and links to paper/code

README.md

@@ -1,19 +1,30 @@
 ---
+datasets:
+- kitti
 library_name: pytorch
-tags:
-  - deltatok
 license: apache-2.0
-datasets:
-  - kitti
+pipeline_tag: depth-estimation
+tags:
+- deltatok
 ---
 
 # Depth Head — KITTI
 
-Monocular depth estimation head trained on KITTI. Requires a frozen [DINOv3](https://github.com/facebookresearch/dinov3) ViT-B backbone (not included).
+Monocular depth estimation head trained on KITTI, as presented in [A Frame is Worth One Token: Efficient Generative World Modeling with Delta Tokens](https://huggingface.co/papers/2604.04913).
+
+Requires a frozen [DINOv3](https://github.com/facebookresearch/dinov3) ViT-B backbone (not included).
+
+- **Project Page:** [https://deltatok.github.io](https://deltatok.github.io)
+- **Code:** [https://github.com/amazon-far/deltatok](https://github.com/amazon-far/deltatok)
 
 ## Usage
 
-See the [DeltaTok GitHub repository](https://github.com/amazon-far/deltatok) for training and evaluation code.
+See the [DeltaTok GitHub repository](https://github.com/amazon-far/deltatok) for training and evaluation code. Evaluation typically involves using the `main.py` script provided in the repository:
+
+```bash
+python main.py validate -c configs/deltatok_vitb_dinov3_vitb_kinetics.yaml \
+  --model.ckpt_path=path/to/deltatok-kinetics/pytorch_model.bin
+```
 
 ## Acknowledgements
 
@@ -29,4 +40,4 @@ See the [DeltaTok GitHub repository](https://github.com/amazon-far/deltatok) for
   booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
   year      = {2026}
 }
-```
+```