GitHub - ale256/representation_unlearning: Code for CVPR 2026 paper, POUR: A Provably Optimal Method for Unlearning Representations via Neural Collapse (https://arxiv.org/abs/2511.19339) and ICML 2026 CATS Workshop paper, DECAF: De-Clustering for Adaptive Representational Unlearning (https://arxiv.org/abs/2607.23934)

Name		Name	Last commit message	Last commit date
cka_eval.py		cka_eval.py
eval_mia.py		eval_mia.py
get_results.py		get_results.py
methods.py		methods.py
metrics.py		metrics.py
models.py		models.py
train.py		train.py
train_subset.py		train_subset.py
train_vit.py		train_vit.py
unlearn.py		unlearn.py
unlearn_vit_pathmnist.py		unlearn_vit_pathmnist.py
utils.py		utils.py

About

Code for CVPR 2026 paper, POUR: A Provably Optimal Method for Unlearning Representations via Neural Collapse (https://arxiv.org/abs/2511.19339) and ICML 2026 CATS Workshop paper, DECAF: De-Clustering for Adaptive Representational Unlearning (https://arxiv.org/abs/2607.23934)

Activity

2 stars

0 watching

0 forks

Report repository

Navigation Menu

About

Releases

Packages

Contributors

Languages

Latest commit

History

Folders and files

About

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages