AI Paper Atlas

Research map / Federated learning and privacy / Communication-efficient FL

H-SFP: Hierarchical Federated Learning with Decoupled Split-Model Prototyping

Trung-Dung Tran, Nguyen Ha, Minh-Duong Nguyen, Van-Dinh Nguyen, Kok-Seng Wong

ECCV 2026 · Federated learning and privacy › Communication-efficient FL · approach: Efficient local training in FL · 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 contrastive loss (ablation)HSFL baseline · CIFAR-100 IIDaccuracy (%): 35.47 -> 41.20Improved
add synthetic data (ablation)HSFL + contrastive loss · CIFAR-100 IIDaccuracy (%): 41.20 -> 47.60Improved
add dimension-wise variance to mean-only statistics (ablation)H-SFP · CIFAR-100 IIDaccuracy (%): 52.35 -> 55.10Improved
add dimension-wise variance to mean-only statistics (ablation)H-SFP · ImageNet-1K IIDaccuracy (%): 26.8 -> 27.9Improved
add model averaging to purely generative variant (ablation)H-SFP · CIFAR-100 IIDaccuracy (%): 48.20 -> 55.10Improved
set contrastive temperature tau 0.5 instead of 0.2 (ablation)H-SFP · CIFAR-100 IIDaccuracy (%): 48.15 -> 55.10Improved
set contrastive temperature tau 0.5 instead of 2.0 (ablation)H-SFP · CIFAR-100 IIDaccuracy (%): 48.65 -> 55.10Improved
use less frequent aggregation intervals (C instead of A) (ablation)H-SFP · CIFAR-10 IIDaccuracy (%): 67.44 -> 65.44Hurt
increase clients from 20 to 200 (data sparsity)FedAvg · CIFAR-100 IIDaccuracy (%): FedAvg 60.50 -> 48.85Hurt
replace SplitFed with H-SFPSplitFed · ISIC-2018training time (h): 72.4 -> 6.8Improved
replace FedAvg with H-SFP (A)FedAvg · CIFAR-10 dir(0.3)accuracy (%): 12.15 -> 40.15Improved
replace FedAvg with H-SFP (C) in communicationFedAvg · CIFAR-100 communicationtotal GB: 353x reduction vs FedAvg; 10.94 GBImproved

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