AI Paper Atlas

Research map / Continual learning / Class-incremental

Selective Amnesia using Contrastive Subnet Erasure for Class Level Unlearning in Vision Models

Vishal Pramanik, Maisha Maliha, Susmit Jha, Alvaro Velasquez, Olivera Kotevska, Sumit Kumar Jha

CVPR 2026 · Continual learning › Class-incremental · approach: Incremental object detection with replay

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
use a non-target set with semantic overlap (bird, ship) instead of none (ablation)CSE · CIFAR-10 forget airplaneretain-test accuracy: 0.76 -> 0.93Improved
use semantically similar non-target classes instead of dissimilar ones (ablation)CSE · CIFAR-10 forget airplaneretain-test accuracy: 0.81 -> 0.93Improved
raise non-target samples from 5% to 10% per category (ablation)CSE · CIFAR-10 forget airplaneretain-test accuracy: 0.85 -> 0.93Improved
raise non-target samples from 10% to 15% per category (ablation)CSE · CIFAR-10 forget airplaneretain-test accuracy: 0.93 -> 0.94No clear effect
training-free contrastive channel attenuation instead of ESC global projectionCSE · CIFAR-10 single-class, ResNet-18forget-test accuracy: .10 -> .01Improved
contrastive channel attenuation instead of DELETE fine-tuningCSE · ImageNet single-class, ResNet-18H-Mean: .65 -> .73Improved
increase forgotten classes from 2 to 5CSE · CIFAR-100 multi-classforget-test accuracy: .02 -> .04No clear effect
increase forgotten classes from 2 to 5ESC · CIFAR-100 multi-classH-Mean: .80 -> .75Hurt
apply CSE to a face model to forget one LFW identityCSE · LFW, ResNet-18target identity verification accuracy: 0.91 -> 0.04Improved
apply CSE to a face model to forget one LFW identityCSE · LFW, ResNet-18non-target identity accuracy: 0.88 -> 0.86No clear effect

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