AI Paper Atlas

Research map / Federated learning and privacy / Personalized and heterogeneous FL

Prevention over Correction: Learning Aligned Representations in One-shot Federated Learning

YongHoon Kang, Jee-Hyong LEE

ECCV 2026 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Federated learning with pretrained models · 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 DMA, VS and feature-level ensemble together (ablation)one-shot FL baseline · CIFAR-10 (alpha=0.05)accuracy: 41.46 -> 69.21Improved
add DMA, VS and feature-level ensemble together (ablation)one-shot FL baseline · CIFAR-100 (alpha=0.05)accuracy: 20.46 -> 41.57Improved
add Variance Stabilizer on top of DMA (ablation)FACE · ablationaccuracyImproved
add feature-level ensemble alone (ablation)baseline · ablationaccuracyNo clear effect
align class means to fixed reference matrix instead of learnable prototypes (FACE vs FAFI)one-shot FL · CIFAR-10 (alpha=0.05)accuracy: 60.28 -> 69.21Improved
FACE vs FuseFL at mild heterogeneityone-shot FL · CIFAR-10 (alpha=0.5)accuracy: 84.34 -> 83.97Hurt
FACE vs FAFI at mild heterogeneityone-shot FL · CIFAR-100 (alpha=0.5)accuracy: 54.42 -> 53.47Hurt
increase clients from 5 to 100 with FACEFACE · CIFAR-10 (alpha=0.5)accuracy: 83.97 -> 77.18Mixed
increase clients from 5 to 100 with FuseFLFuseFL · CIFAR-10 (alpha=0.5)accuracy: 84.34 -> 37.11Hurt
FACE vs FAFI at tiny-scale heterogeneity extremeone-shot FL · Tiny-ImageNet (alpha=0.05)accuracy: 34.60 -> 39.20Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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