AI Paper Atlas

Research map / Continual learning / Class-incremental

From Selection to Scheduling: Federated Geometry-Aware Correction Makes Exemplar Replay Work Better under Continual Dynamic Heterogeneity

Zhuang Qi, Ying-Peng Tang, Lei Meng, Guoqing Chao, Lei Wu, Han Yu, Xiangxu Meng

CVPR 2026 · Continual learning › Class-incremental · approach: Class-balanced knowledge retention · 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 fixed ETF classifier alone (ablation)FedCBDR · CIFAR10 3 tasks beta=0.5, K=5Top-1 accuracy (%): 63.32 -> 62.17Hurt
add fixed ETF classifier alone (ablation)FedCBDR · CIFAR100 5 tasks beta=0.1, K=5Top-1 accuracy (%): 45.84 -> 44.58Hurt
add GSA on top of ETF (ablation)FedCBDR+ETF · CIFAR10 3 tasks beta=0.5, K=5Top-1 accuracy (%): 62.17 -> 68.54Improved
add EGC on top of ETF (ablation)FedCBDR+ETF · CIFAR10 3 tasks beta=0.5, K=5Top-1 accuracy (%): 62.17 -> 69.12Improved
add EGC on top of ETF+GSA (ablation)FedCBDR+ETF+GSA · CIFAR10 3 tasks beta=0.5, K=5Top-1 accuracy (%): 68.54 -> 72.67Improved
add GSA on top of ETF+EGC (ablation)FedCBDR+ETF+EGC · CIFAR100 5 tasks beta=0.1, K=5Top-1 accuracy (%): 47.16 -> 50.14Improved
plug FEAT into FedCBDR (FEAT_F)FedCBDR · CIFAR10 3 tasks beta=0.5, K=5Top-1 accuracy (%): 63.32 -> 72.67Improved
plug FEAT into Re-Fed+ (FEAT_R)Re-Fed+ · CIFAR100 5 tasks beta=0.1, K=5Top-1 accuracy (%): 31.92 -> 37.14Improved
add FedRCL to FedCBDRFedCBDR · CIFAR10 5 tasks beta=0.5, K=5Top-1 accuracy (%): 61.77 -> 58.74Hurt
add MOON to FedCBDRFedCBDR · CIFAR100 5 tasks beta=0.1, K=5Top-1 accuracy (%): 45.84 -> 45.14No clear effect
add FEAT to FedCBDR instead of FedFSAFedCBDR · CIFAR10 5 tasks beta=0.5, K=5Top-1 accuracy (%): 64.42 -> 70.19Improved
add GSA and EGC to FedCBDR, last task accuracy on CIFAR10FedCBDR · CIFAR10 3 tasks beta=0.5, Task 3 accuracyTop-1 accuracy (%): 96.80 -> 94.28Hurt

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