Add paper, project page, and code links (#2)

- Add paper, project page, and code links (f4157b192989bf54418c5c446752f292cbed72d3)


Co-authored-by: Niels Rogge <nielsr@users.noreply.huggingface.co>

README.md

@@ -2,12 +2,21 @@
 license: cc-by-nc-4.0
 size_categories:
 - 10K<n<100K
+task_categories:
+- other
 ---
 
-## FID Evaluation Artifacts
+# Sphere Encoder FID Evaluation Artifacts
 
-Download the evaluation artifacts for the [**Sphere Encoder**](https://github.com/facebookresearch/sphere-encoder) project and place them in `./workspace/`. 
-The directory tree should look like this:
+This repository contains the evaluation artifacts for the paper [Image Generation with a Sphere Encoder](https://huggingface.co/papers/2602.15030).
+
+[**Project Page**](https://sphere-encoder.github.io) | [**GitHub Repository**](https://github.com/facebookresearch/sphere-encoder)
+
+These artifacts include data statistic files (`fid_stats`) and reference images (`fid_refs`) used to calculate Fréchet Inception Distance (FID) for generative models across several datasets, including CIFAR-10, ImageNet, Animal Faces, and Oxford Flowers.
+
+## Workspace Setup
+
+Download the evaluation artifacts and place them in your `./workspace/` directory. The directory tree should look like this:
 
 ```bash
 ./workspace/
@@ -20,3 +29,29 @@ The directory tree should look like this:
     |── ref_images_imagenet_256px/images
 ```
 
+## Sample Usage
+
+To evaluate a trained model using these artifacts, you can use the evaluation script provided in the [GitHub repository](https://github.com/facebookresearch/sphere-encoder):
+
+```bash
+./run.sh eval.py \
+  --job_dir sphere-base-base-cifar-10-32px \
+  --forward_steps 1 4 \
+  --report_fid rfid gfid \
+  --use_cfg True \
+  --cfg_min 1.2 \
+  --cfg_max 1.2 \
+  --cfg_position combo \
+  --rm_folder_after_eval True
+```
+
+## Citation
+
+```bibtex
+@article{yue2025image,
+  title={Image Generation with a Sphere Encoder},
+  author={Yue, Kaiyu and Jia, Menglin and Hou, Ji and Goldstein, Tom},
+  journal={arXiv preprint arXiv:2602.15030},
+  year={2025}
+}
+```