AI Paper Atlas

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

Trustworthy Federated Label Distribution Learning under Annotation Quality Disparity

Junxiang Wu, Zhiqiang Kou, Hongwei Zeng, Wenke Huang, Biao Liu, Hanlin Gu, Yuheng Jia, Di Jiang, Yang Liu, Xin Geng

ICML 2026 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Federated semi-supervised and graph 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 client-side global-semantic-anchor rectifier to FedAvg (ablation)FedAvg · FER-LDLKL divergence: 0.342 -> 0.264Improved
add client-side anchor rectifier to FedAvg (ablation)FedAvg · PIPAL-LDLKL divergence: 0.252 -> 0.154Improved
add server-side reliability-aware aggregation on top of anchor rectifier (ablation)FedQual (+A) · PIPAL-LDLKL divergence: 0.154 -> 0.116Improved
add server-side reliability-aware aggregation on top of anchor rectifier (ablation)FedQual (+A) · FER-LDLKL divergence: 0.264 -> 0.253Improved
add anchor rectifier only (without aggregation change) on PIPAL-LDL (ablation)FedAvg · PIPAL-LDLClark distance: 1.398 -> 1.399No clear effect
use full FedQual instead of FedAvg (ablation)FedAvg · PIPAL-LDLCosine similarity: 0.929 -> 0.958Improved
replace sample-size weighting with quality-weighted aggregation alone (FedQAgg)FedAvg · PIPAL-LDLKL divergence: 0.2521 -> 0.4398Hurt
replace sample-size weighting with quality-weighted aggregation alone (FedQAgg)FedAvg · FER-LDLKL divergence: 0.3420 -> 0.2756Improved
replace sample-size weighting with quality-weighted aggregation alone (FedQAgg)FedAvg · FI-LDLKL divergence: 0.4233 -> 0.5451Hurt
add FedProx proximal term to FedAvgFedAvg · FI-LDLKL divergence: 0.4233 -> 0.5119Hurt
add MOON contrastive alignment to FedAvgFedAvg · KADID-LDLKL divergence: 0.1248 -> 0.4262Hurt
increase fraction of noisy clients to 75% (ablation)FedQual · four Fed-LDL benchmarksall six LDL metricsNo clear effect

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