AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

Unlearning Isn't Deletion: Investigating Reversibility of Machine Unlearning in LLMs

Xiaoyu Xu, Xiang Yue, Yang Liu, Qingqing Ye, Huadi Zheng, Peizhao Hu, Minxin Du, Haibo Hu

ICML 2026 · Large language models › Language, safety and interpretability · approach: LLM pruning and scaling data · 0 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
relearn briefly on the forget set after single unlearningGA, GA+KL, RLabel unlearning · Single unlearning, simple task (Yi-6B)forget accuracy: GA+KL and RLabel recover F.Acc to approximately 77%Improved
relearn on the cumulative forget set after continual unlearning at moderate LR (5e-6, N=100)GA continual unlearning · Simple task (Yi-6B)F.Acc: 9.1 -> 80.0Improved
relearn after aggressive continual unlearning (LR=3e-5, N=100)GA continual unlearning · Simple task (Yi-6B)F.Acc: 0.0 -> 2.1No clear effect
increase learning rate of continual unlearning from 5e-6 to 3e-5 (ablation)GA continual unlearning · Mean PCA distance on forget set after relearning (Yi-6B)mean PCA distance: 1.37 -> 104.58Hurt
add a retain-set loss (GD or KL) to GA or NPO in single unlearning (ablation)GA / NPO unlearning · Single unlearning, simple task, LR=3e-6 (Yi-6B)F.Acc after unlearning: GA 65.4; GA+GD 75.1; GA+KL 71.0; NPO 76.2; NPO+KL 76.3Mixed
use mean PCA distance before relearning as predictor of recoverabilityDiagnostic toolkit · GA continual unlearning (Yi-6B)recovery of F.Acc: 11.52 vs 133.20 mean PCA distance (forget) pre-relearningImproved
relearning on forget set after continual unlearning pushes accuracy above the originalGA, NPO, NPO+KL, GA+GD · Continual unlearning, simple task (Yi-6B)F.Acc: 78.9 -> 83.8 (NPO+KL, LR=5e-6, N=100)Improved
min-k%-prob MIA AUC as evaluation of unlearningMIA metric · Continual unlearning GA, LR=3e-5 (Yi-6B)MIA AUC: 26.1 -> 74.5 after relearning while F.Acc is 0.0 -> 2.1Hurt
use GA+GD+WAGLE with relearning restricted to 50% retain setGA+GD+WAGLE · Yi-6B, LR=2e-5, N=50F.Acc / R.Acc after relearning: F.Acc 37.8 -> 46.9; R.Acc 55.9 -> 58.3No clear effect
use GA+GD+WAGLE with relearning on unrelated dataGA+GD+WAGLE · Yi-6B, LR=4e-5, N=50F.Acc after relearning: 27.8 -> 31.5No clear effect
recover forgotten knowledge via prompt attacks, jailbreaking, quantization or in-context recovery instead of relearning (ablation)Unlearned models · Reversible catastrophic forgetting regimerecovery of forgotten knowledgeHurt
relearn on forget set vs retain or unrelated data (ablation)Relearning probe · Sample efficiency (Appendix)recovery speedImproved

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