AI Paper Atlas

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

FedRG: Unleashing the Representation Geometry for Federated Learning with Noisy Clients

Tian Wen, Zhiqin Yang, Yonggang Zhang, Xuefeng Jiang, Hao Peng, Yuwei Wang, Bo Han

CVPR 2026 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Federated semi-supervised and graph learning · 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 noise absorption matrix (ablation)FedRG · CIFAR-10 (4 noise scenarios)Clean recall, CRA, accuracyImproved
Add label-decoupled spherical (SimCLR) pretraining stage (ablation)FedRG · CIFAR-10 (4 noise scenarios)Accuracy, CRAImproved
Use symmetric cross-entropy instead of standard CE (ablation)FedRG · CIFAR-10 (4 noise scenarios)AccuracyImproved
Keep noise absorption matrix local (do not aggregate) (ablation)FedRG · CIFAR-10 (4 noise scenarios)AccuracyImproved
Use FedRG instead of best baselineSymmetricCE · CIFAR-10 symmetric globalizedAccuracy: 59.23 -> 63.29Improved
Use FedRG instead of FedSAMFedSAM · SVHN pairflip globalizedAccuracy: 75.25 -> 80.62Improved
Use FedRG instead of SymmetricCESymmetricCE · CIFAR-100 symmetric globalizedAccuracyMixed
Increase number of semantic clusters (6 to 20) (ablation)FedRG · CIFAR-10 symmetric globalizedAccuracy: 62.60 - 64.00No clear effect
Increase proportion of selected clients per round (ablation)FedRG · CIFAR-10 symmetric globalizedAccuracyImproved

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