AI Paper Atlas

Research map / Continual learning / Continual fine-tuning of LLMs

Redirection for Erasing Memory (REM): Towards a universal unlearning method for corrupted data

Stefan Schoepf, Michael Curtis Mozer, Nicole Elyse Mitchell, Alexandra Brintrup, Georgios Kaissis, Peter Kairouz, Eleni Triantafillou

ICLR 2026 · Continual learning › Continual fine-tuning of LLMs · approach: Machine unlearning by selective forgetting · 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
add the redirection step 3.1 (train on full data with shared mask for forget set) (ablation)REM · CIFAR10 ResNet-9 (50% capacity)Utility x Healed: 69.82 ± 1.85 -> 73.40 ± 1.43Improved
add the barrier step 3.2 (NPO on theta_o1 during repair) (ablation)REM without ETD · CIFAR10 ResNet-9 (50% capacity)Utility x Healed: 71.38 ± 1.47 -> 73.40 ± 1.43Improved
add the barrier step 3.2 (ablation)REM with ETD · CIFAR10 ResNet-9 (50% capacity)Utility x Healed: 73.69 ± 0.69 -> 73.19 ± 0.72No clear effect
train the original model with ETD (ablation)REM · CIFAR10 ResNet-9 (50% capacity)Utility x Healed: 73.40 ± 1.43 -> 73.19 ± 0.72No clear effect
use REM instead of NPO aloneNPO · CIFAR10 ResNet-9 (50% capacity)Utility x Healed: 56.40 ± 2.50 -> 73.40 ± 1.43Improved
add ETD pretrainingNPO · CIFAR10 ResNet-9 (50% capacity)Utility x Healed: 56.40 ± 2.50 -> 67.10 ± 1.17Improved
add ETD pretrainingSCRUB · CIFAR10 ResNet-9 (50% capacity)Utility x Healed: 42.69 ± 2.33 -> 59.85 ± 2.50Improved
use 100% capacity model instead of 50% capacityREM · CIFAR10 ResNet-9 (50% capacity)Utility x Healed: 73.40 ± 1.43 -> 59.65 ± 2.99Hurt
use REM instead of Ascent on ViT SVHNAscent · SVHN ViT (50% capacity)Utility x Healed: 67.14 ± 0.89 -> 73.27 ± 0.32Improved
use retraining from scratch on retain setRetrained · CIFAR10 ResNet-9 (50% capacity)Utility x Healed: 48.52 ± 2.47 vs No UL+ETD 50.11 ± 2.75Hurt
use REM instead of SCRUB on ImagenetteSCRUB · Imagenette (100 corruptions)Utility x Healed: 54.52 ± 2.89 -> 60.98 ± 2.02Improved
lower capacity model or ETD for high regularity corruptionsResNet-9 · poison trigger (high regularity)HealedNo 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