AI Paper Atlas

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

Federated Distillation for Whole Slide Image via Gaussian-Mixture Feature Alignment and Curriculum Integration

Luru Jing, Cong Cong, Yanyuan Chen, Yongzhi Cao

ICML 2026 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Federated medical and multimodal 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 feature-level distillation (FDD) instead of pixel-level (ablation)FedHD · CAMELYON16 / CAMELYON17 / TCGA-IDHaccuracy (%): 88.7 -> 89.7 / 77.2 -> 78.8 / 80.5 -> 82.1Improved
add Gaussian-mixture alignment (GMA) alone (ablation)FedHD · CAMELYON16accuracy (%): 88.7 -> 89.4Improved
add Gaussian-mixture alignment (GMA) alone (ablation)FedHD · TCGA-IDHaccuracy (%): 80.5 -> 83.4Improved
add one-to-one distillation (O2O) to FDD (ablation)FedHD · CAMELYON17accuracy (%): 78.8 -> 81.2Improved
add GMA on top of FDD&O2O (ablation)FedHD · CAMELYON17accuracy (%): 81.2 -> 81.9Improved
add curriculum-based federation (CBF) alone (ablation)FedHD · CAMELYON17accuracy (%): 77.2 -> 80.7Improved
add CBF on top of FDD&O2O&GMA (ablation)FedHD · CAMELYON16accuracy (%): 91.2 -> 91.2No clear effect
add CBF on top of FDD&O2O&GMA (ablation)FedHD · TCGA-IDHaccuracy (%): 84.6 -> 84.8Improved
FedHD vs strongest baseline FedWSIDD on heterogeneous modelsFedHD · CAMELYON16accuracy (%): 88.7 -> 91.2Improved
FedHD vs FedWSIDD privacy under membership inferenceFedHD · TCGA-IDHLiRA mean AUC (%): 56.1 -> 52.3Improved
feature-level distillation instead of image distillationFedHD · TCGA-IDH client C1distillation time: 10-12 hours -> approximately 1 hourImproved
enable O2O distillationFedHD · TCGA-IDH client C1communication payload: approximately 39 MB -> approximately 1.19 GBHurt

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