DECAF: De-Clustering for Adaptive Representational Unlearning

July 27, 2026 ยท Grace Period ยท ๐Ÿ› ICML 2026 CATS Workshop

โณ Grace Period
This paper is less than 90 days old. We give authors time to release their code before passing judgment.
Authors Anjie Le, Can Peng, Hongcheng Guo, J. Alison Noble arXiv ID 2607.23934 Category cs.LG: Machine Learning Citations 0 Venue ICML 2026 CATS Workshop
Abstract
Machine unlearning, which aims to remove the influence of specific training data from a trained model, is a key requirement for privacy, accountability, and adaptive deployment. We argue that many unlearning methods are vulnerable to a simple clustering attack, which can recover class structure in an unsupervised manner, limiting their suitability for continual deployment where removal requests must be handled reliably on demand. To address this, we propose DECAF (DE-Clustering for Adaptive Forgetting), a post-hoc method that operates only on the forget set and is designed to break the cluster. DECAF combines input noise, confidence suppression, and entropy-based output diversification to disrupt the residual feature-space structure associated with forgotten data. On CIFAR-10 with ResNet-18, DECAF attains 0.10% forget-class accuracy, 79.4% retain accuracy, and an AUS of 0.88, surpassing all other baselines. In cluster-based analysis, it attains performance comparable to that of unlearning methods that use the full training set, while being significantly more efficient. Code: https://github.com/ale256/representation_unlearning.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

๐Ÿ“œ Similar Papers

In the same crypt โ€” Machine Learning