AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Unlearning during Training: Domain-Specific Gradient Ascent for Domain Generalization

Di Zhao, Jingfeng Zhang, Hongsheng Hu, Philippe Fournier-Viger, Gillian Dobbie, Yun Sing Koh

ICLR 2026 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Optimization-based domain generalization

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 post-epoch Identify and Unlearn (IU)ERM · PACSaccuracy (%): 83.0 -> 85.7Improved
add EMA smoothing to the unlearning score (ablation)ERM + IU · DomainNetaccuracy (%): 42.2 -> 43.1Improved
add IU to strong baselinesUDIM / VL2V · PACSaccuracy (%): 84.7 -> 86.6 (UDIM); 85.5 -> 87.0 (VL2V)Improved
apply gradient ascent on all parameters using unlearning set (no channel selection) (ablation)ERM · PACSaccuracy (%): 83.0 -> 78.9Hurt
apply gradient ascent on domain-specific channels over entire training set (no unlearning set selection) (ablation)ERM · PACSaccuracy (%): 83.0 -> 76.7Hurt
combine unlearning set selection with domain-specific channel selection (ablation)ERM · PACSaccuracy (%): 78.9 -> 85.7Improved
use Inter-Domain Variance instead of Aggregated Variance for channel selection (ablation)ERM + IU · PACSaccuracy (%): 84.3 -> 85.7Improved
remove unlearning set from training data and retrain (instead of unlearning) (ablation)ERM · PACSaccuracy (%): 83.0 -> 80.1Hurt
set alpha away from 1.1 (ablation)ERM + IU · PACSaccuracy (%): 85.7 -> 83.6 (0.8); 84.1 (1.3)Mixed
vary MAD threshold k from 2.0 (ablation)ERM + IU · PACSaccuracy (%): 85.7 -> 83.4 (k=1.0); 83.3 (k=3.0)Mixed
use MAD thresholding instead of fixed bottom-10% selection (ablation)unlearning set selection · PACSadaptivity of set sizeImproved

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