AI Paper Atlas

Research map / Continual learning / Class-incremental

POUR: A Provably Optimal Method for Unlearning Representation via Neural Collapse

Anjie Le, Can Peng, Yuyuan Liu, J. Alison Noble

CVPR 2026 · Continual learning › Class-incremental · approach: Class-balanced knowledge retention

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
distill feature extractor on forget set toward projected features (POUR-D) instead of post-hoc projection only (POUR-P) (ablation)POUR · CIFAR-10 (ResNet18)Acc_r: 94.97 -> 92.86Hurt
use POUR-D to modify encoder representations (POUR-P leaves encoder unchanged) (ablation)POUR · CIFAR-10 (ResNet18)rMIA: 56.67 -> 51.80Improved
POUR-D versus best forget-set-only baselines on representation metricsPOUR-D · CIFAR-10 (ResNet18)RUS(o): 0.80 -> 0.85Improved
POUR-D versus DELETE on retain accuracyPOUR-D · CIFAR-10 (ResNet18)Acc_r: 88.73 -> 92.86Improved
POUR-P versus Gradient Ascent on CIFAR-100POUR-P · CIFAR-100 (ResNet18)Acc_r: 50.46 -> 77.65Improved
use POUR-D instead of POUR-P on CIFAR-100 (ablation)POUR · CIFAR-100 (ResNet18)Acc_r: 77.65 -> 73.44Hurt
weak class separation (CIFAR-100 entangled classes) makes forget-set-only unlearning harderforget-set-only unlearning · CIFAR-100 (ResNet18)RUS(r): 0.65 (POUR-D) vs 0.84 on CIFAR-10 RUS(r) not reported; CIFAR-10 POUR-D 0.47Hurt
apply POUR to pretrained ViT with domain shiftPOUR-P · PathMNIST internal vs externalAcc_r: 87.14 internal, 87.44 externalImproved
Random Label and Gradient Ascent on PathMNISTbaselines · PathMNISTAcc_r internal vs external: 81.43 -> 76.72 (Gradient Ascent)Hurt
POUR-P on CLIP text-embedding removal for ImageNet classesPOUR-P · ImageNet (CLIP-L/14)Acc_f: 94.00 -> 14.00 (Goldfish)Improved
POUR on semantic segmentation forgettingDeepLabV3+ · VOC 2012IoU_f and IoU_r: Dog IoU_f 90.98 -> 0.00; IoU_r 77.81 -> 71.53Mixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy