AI Paper Atlas

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

FedCART: Tackling Long-Tailed Distributions in Federated Adversarial Training via Classifier Refinement

Yuchen Qin, Yizhi Zhou, Junxiao Wang, Xin Xie, Heng Qi

CVPR 2026 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Federated learning with pretrained models

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 alignment loss LAlign between natural and adversarial features (ablation)FedCART · CIFAR10-LTrobust accuracy PGD-20: 29.33 -> 32.57Improved
add adversarial loss LAT (ablation)FedCART · CIFAR10-LTrobust accuracy PGD-20: 2.91 -> 32.57Improved
add server-side classifier re-training with auxiliary classifier (ablation)FedCART · CIFAR10-LTnatural accuracy: 49.98 -> 67.24Improved
replace CalFAT with FedCART under long-tailed dataCalFAT · CIFAR10-LTrobust accuracy AA: 19.93 -> 27.47Improved
FedCART vs CalFAT under small beta (strong heterogeneity)FedCART · CIFAR10-LT (beta=0.25)natural accuracy: 55.65 -> 51.52Hurt
FedCART vs CalFAT under small beta (strong heterogeneity)FedCART · CIFAR10-LT (beta=0.25)robust accuracy AA: 18.00 -> 24.72Improved
FedCART vs CalFAT on FMNIST-LT natural accuracyFedCART · FMNIST-LTnatural accuracy: 86.75 -> 85.13Hurt
increase imbalance ratio rhoCalFAT and FedCART · CIFAR10-LTaccuracyHurt
combine TRADES with FedCARTFedTRADES · CIFAR10-LTnatural accuracy: 41.59 -> 65.48Improved
combine TAET with FedCARTFedTAET · CIFAR10-LTrobust accuracy PGD-20: 22.42 -> 26.57Improved
FedCART on majority classes vs FedFATFedCART · CIFAR100-LT majority classesnatural accuracy: 60.45 -> 52.16Hurt
decrease client active rate or increase client countCalFAT and FedCART · CIFAR10-LTaccuracyHurt

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