Add model citation and fine-tuning dataset note

README.md

@@ -157,9 +157,11 @@ pose_labels/
 
 or as a CSV with `image_path, pose` columns. Class counts are inherently imbalanced and are handled at the sampler level (see below).
 
-More detail is provided in the [training dataset repository](https://huggingface.co/datasets/imageomics/mmla-pose).
+This model was fine-tuned on the MMLA pose dataset:
 
-[More Information Needed — exact per-class sample counts, splits, and dataset card link]
+- Dataset: [imageomics/mmla-pose](https://huggingface.co/datasets/imageomics/mmla-pose)
+
+The dataset contains cropped images of zebras from MMLA drone footage labeled with one of eight pose orientations: front, front-left, front-right, left, back-left, back, back-right, and right.
 
 ### Training Procedure
 
@@ -303,27 +305,34 @@ Training objective: cross-entropy with label smoothing (0.1), optimized only ove
 
 ## Citation
 
-[More Information Needed]
-
-<!--
-If you use our model in your work, please cite the model and any associated paper.
+If you use this model, please cite this model repository, the MMLA pose dataset, the associated CV4Animals workshop paper, and the underlying DINOv2 backbone.
 
 **Model**
-```
-@software{<ref_code>,
-  author = {<author1 and author2>},
-  doi = {<doi once generated>},
-  title = {DINOv2 8-Class Animal Pose Classifier},
-  version = {<version#>},
+
+```bibtex
+@software{imageomics_mmla_dino_pose_2026,
+  author = {Sun, Claire and Kline, Jenna and Pillai, Bharath and Berger-Wolf, Tanya},
+  title = {MMLA DINOv2 8-Class Animal Pose Classifier},
   year = {2026},
-  url = {https://huggingface.co/imageomics/mmla-dino-pose}
+  url = {https://huggingface.co/imageomics/mmla-dino-pose},
+  note = {Fine-tuned on the MMLA pose dataset: https://huggingface.co/datasets/imageomics/mmla-pose}
 }
 ```
--->
 
-Underlying backbone:
+**Dataset**
 
+Please also cite the MMLA pose dataset:
+```bibtex
+@dataset{imageomics_mmla_pose_2026,
+  title = {MMLA Pose Dataset},
+  year = {2026},
+  url = {https://huggingface.co/datasets/imageomics/mmla-pose}
+}
 ```
+
+**Underlying backbone**
+
+```bibtex
 @article{oquab2023dinov2,
   title   = {DINOv2: Learning Robust Visual Features without Supervision},
   author  = {Oquab, Maxime and Darcet, Timoth{\'e}e and Moutakanni, Th{\'e}o and Vo, Huy V. and Szafraniec, Marc and Khalidov, Vasil and Fernandez, Pierre and Haziza, Daniel and Massa, Francisco and El-Nouby, Alaaeldin and others},