AI Paper Atlas

Research map / Learning theory and algorithms / Optimal transport

Dataset Distillation via the Wasserstein Metric

Haoyang Liu, Yijiang Li, Tiancheng Xing, Peiran Wang, Vibhu Dalal, Luwei Li, Jingrui He, Haohan Wang

ICCV 2025 · Learning theory and algorithms › Optimal transport · approach: Wasserstein barycenters and GW · 34 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 Per-Class BatchNorm (PCBN) instead of standard BN (ablation)WMDD (Wasserstein loss) · ImageNette 10 IPCtop-1 accuracy: 60.7 -> 64.7Improved
add PCBN instead of standard BN (ablation)WMDD (Wasserstein loss) · ImageNet-1K 10 IPCtop-1 accuracy: 26.8 -> 38.1Improved
add PCBN instead of standard BN (ablation)CE feature loss · ImageNette 10 IPCtop-1 accuracy: 54.2 -> 63.5Improved
add PCBN instead of standard BN (ablation)CE feature loss · Tiny ImageNet 10 IPCtop-1 accuracy: 38.0 -> 41.0Improved
use Wasserstein barycenter loss instead of cross-entropy (ablation)WMDD with PCBN · ImageNet-1K 10 IPCtop-1 accuracy: 36.4 -> 38.1Improved
use Wasserstein barycenter loss instead of cross-entropy (ablation)WMDD with PCBN · ImageNette 10 IPCtop-1 accuracy: 63.5 -> 64.7Improved
use Wasserstein loss instead of cross-entropy with standard BN (ablation)WMDD with BN · Tiny ImageNet 10 IPCtop-1 accuracy: 38.0 -> 36.6Hurt
replace Wasserstein metric with sliced Wasserstein (ablation)WMDD · Tiny ImageNet 50 IPCtop-1 accuracy: 59.4 -> 58.3No clear effect
use features from layer 18 instead of layer 5 (ablation)WMDD · Tiny ImageNettop-1 accuracy: 2.4 -> 41.8Improved
replace Wasserstein metric with MMD (ablation)WMDD · Tiny ImageNet and ImageNet-1Ktop-1 accuracyHurt
use WMDD instead of CDA for ViT-Tiny evaluationWMDD · ImageNet-1K 50 IPC cross-architecturetop-1 accuracy: 31.22 -> 34.25Improved

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