Clustering via Self‑Supervised Diffusion

How to use

Image embeddings

For generating image embeddings for different pretrained models and datasets, we follow the TEMI pipeline from
TEMI-official-BMVC2023.

Concretely, we rely on their gen_embeds.py script to extract features for a given architecture and dataset, e.g.:

python gen_embeds.py --arch clip_ViT-B/32 --dataset CIFAR10 --batch_size 256

python cludi.py --num_clusters 50

---

## Citation If you find this work useful, please cite:

@inproceedings{uziel2025clustering,
  title     = {Clustering via Self-Supervised Diffusion},
  author    = {Uziel, Roy and Chelly, Irit and Freifeld, Oren and Pakman, Ari},
  booktitle = {International Conference on Machine Learning},
  year      = {2025},
}