AI Paper Atlas

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

FedMC: Federated Manifold Calibration

Yanbiao Ma, Wei Dai, Gaoyang Jiang, wanyi Chen, Chenyue Zhou, Yiwei Zhang, Fei Luo, Junhao Wang, Andi Zhang

ICLR 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 FedMC manifold calibrationFedVTP · Tiny-ImageNet (beta=0.1)accuracy: 77.58 -> 80.12Improved
add FedMC manifold calibrationFedVTP · CIFAR-100 (beta=0.1)accuracy: 81.01 -> 85.08Improved
replace global linear GGEUR with nonlinear manifold FedMC (ablation)FedVTP · CIFAR-100 (beta=0.1)accuracy: 82.55 -> 85.08Improved
replace global linear GGEUR with FedMC (ablation)FedVTP · Tiny-ImageNet (beta=0.5)accuracy: 81.15 -> 81.53Improved
add FedMC to FedVTPFedVTP · DomainNet (beta=0.1)accuracy: 83.82 -> 85.93Improved
replace GGEUR with FedMC (ablation)FedVTP · DomainNet (beta=0.1)accuracy: 83.85 -> 85.93Improved
replace GGEUR with FedMC (ablation)FedVTP · Office-Home (beta=0.1)accuracy: 89.15 -> 91.03Improved
add FedMC to FedProtoFedProto · Office-Home-LDS (beta=0.1)accuracy: 69.40 -> 85.92Improved
replace GGEUR with FedMC (ablation)FedOPT · Office-Home-LDS (beta=0.1)accuracy: 84.20 -> 86.58Improved
replace GGEUR with FedMC (ablation)MOON · Office-Home-LDS (beta=0.1)accuracy: 78.08 -> 80.73Improved
add FedMC to FedAvgFedAvg · Office-Home-LDS (beta=0.1)accuracy: 70.14 -> 85.11Improved
add FedMC with 500 clientsFedVTP · CIFAR-100 (beta=0.1, K=500)accuracy: 69.85 -> 73.26Improved

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