AI Paper Atlas

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

ConstStyle: Robust Domain Generalization with Unified Style Transformation

Nam Duong Tran, Nam Nguyen Phuong, Hieu H. Pham, Phi Le Nguyen, My T. Thai

ICCV 2025 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Foundation-model domain generalization · 1 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
use pretrained features instead of learned style statistics for clustering (ablation)ConstStyle · Digits5accuracy: 86.88 -> 76.61Hurt
use pretrained features instead of learned style statistics for clustering (ablation)ConstStyle · PACSaccuracy: 86.77 -> 86.31No clear effect
use GMM clustering instead of domain labels (ablation)ConstStyle · Digits5accuracy: 86.37 -> 86.88Improved
use GMM clustering instead of domain labels (ablation)ConstStyle · PACSaccuracy: 86.73 -> 86.77No clear effect
use partial alignment alpha=0.6 instead of alpha=1.0 (ablation)ConstStyle · PACSaccuracy: 85.86 -> 86.77Improved
use partial alignment alpha=0.5 instead of alpha=0 (full mapping) (ablation)ConstStyle · Digits5accuracy: 85.96 -> 86.88Improved
use 4 clusters instead of 1 (ablation)ConstStyle · Digits5accuracy: 85,80 -> 86.88Improved
use batch size 32 instead of 256 (ablation)ConstStyle · PACSaccuracy: 85.10 -> 86.77Improved
use batch size 32 instead of 8 (ablation)ConstStyle · PACSaccuracy: 85.43 -> 86.77Improved
apply ConstStyle unified-domain alignment vs ERMResNet18 · PACS (single unseen domain)avg accuracy: 79.86 -> 86.77Improved
apply ConstStyle vs ERM at corruption level 5WideResNet · CIFAR10-C level 5accuracy: 47.80 -> 63.63Improved
apply ConstStyle vs ERM at corruption level 1WideResNet · CIFAR10-C level 1accuracy: 87.33 -> 86.59Hurt

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