AI Paper Atlas

Research map / Continual learning / Class-incremental

FG-OrIU: Towards Better Forgetting via Feature-Gradient Orthogonality for Incremental Unlearning

Qian Feng, JiaHang Tu, Mintong Kang, Hanbin Zhao, Chao Zhang, Hui Qian

ICCV 2025 · Continual learning › Class-incremental · approach: Class-balanced knowledge retention · 2 citations (OpenAlex)

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
remove gradient orthogonal projection for remaining classes (add gop-r) (ablation)FG-OrIU · CASIA-Face100 IU Task 1Acc_r: 71.45 -> 74.25Improved
add feature orthogonal projection for forgetting (fop-f) (ablation)FG-OrIU · CASIA-Face100 IU Task 1Acc_f: 0.20 -> 0.00Improved
add gradient orthogonal projection for forgetting (gop-f) (ablation)FG-OrIU · CASIA-Face100 IU Task 2Acc_f: 4.37 -> 0.00Improved
add feature orthogonal projection for remaining (fop-r) (ablation)FG-OrIU · CASIA-Face100 IU Task 2Acc_r: 74.98 -> 75.50Mixed
add feature and gradient orthogonality to LoRA-based incremental unlearning (vs GS-LoRA)GS-LoRA · CASIA-Face100 IU Task 4Acc_r: 75.36 -> 80.51Improved
add feature and gradient orthogonality (vs GS-LoRA) in static unlearningGS-LoRA · CASIA-Face100 static MU (100-90)Acc_r: 73.56 -> 77.15Improved
use FG-OrIU instead of DER++ on ImageNet100 Task 4DER++ · ImageNet100 IU Task 4H-Mean: 89.24 -> 89.15No clear effect
apply FG-OrIU instead of prior methods to hinder feature reconstructionunlearned models · DIP reconstructionSSIM: 0.75 -> 0.01Improved
freeze backbone and retrain head on prior unlearned modelsexisting IU methods · CASIA-Face100 recoveryforgetting-class accuracy recovered: 15%-70% of original performance recoveredHurt

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