AI Paper Atlas

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

Federated Manifold Learning (FML): Tackling Domain Heterogeneity with Structural Knowledge Transfer

Xutong Mu, Yanbiao Ma, Jia Shi, Xueli Geng, Fengkai Xiang, Tao Zhang, Ke Cheng, Yulong Shen

ICML 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 manifold approximation loss only (ablation)FML · DigitsAVG accuracy: 79.44 -> 82.04Improved
add manifold separation loss only (ablation)FML · DigitsAVG accuracy: 79.44 -> 81.24Improved
combine both manifold losses (ablation)FML · DigitsAVG accuracy: 82.04 -> 82.91Improved
add manifold approximation loss only (ablation)FML · Office31AVG accuracy: 50.48 -> 54.75Improved
add manifold separation loss only (ablation)FML · Office31AVG accuracy: 50.48 -> 51.53Improved
combine both manifold losses (ablation)FML · Office31AVG accuracy: 54.75 -> 58.72Improved
replace class prototypes with compressed attention-fused manifoldsFML vs prototype FPL · DigitsAVG accuracy: 80.62 -> 82.91Improved
replace class prototypes with compressed manifoldsFML · Office31AVG accuracy: 52.25 -> 58.72Improved
use MOON contrastive regularizationMOON · DigitsAVG accuracy: 79.44 -> 71.90Hurt
use prototype-based FPL on Office31FPL · Office31AVG accuracy: 50.48 -> 46.83Hurt
raise MML coefficient gamma to 0.9 on Digits (ablation)FML · DigitsAVG accuracy: 82.54 -> 84.14Improved
choose compression size r near 10 (ablation)PMC · Digits and Office31IEIImproved

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