AI Paper Atlas

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

Server-Proximal Aggregation for Federated Domain-Incremental Learning under Partial Participation: Task-Uniform Convergence and Backward Transfer

Longtao Xu, Jian Li

ICML 2026 · Federated learning and privacy › Communication-efficient FL · approach: Client participation analysis · 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
move proximal anchor from client side to server side (ablation)SPECIAL · VLCSACC (%): 51.17 -> 54.92Improved
move proximal anchor from client side to server side (ablation)SPECIAL · PACSACC (%): 43.30 -> 45.29Improved
move proximal anchor from client side to server side (ablation)SPECIAL · DN4ILACC (%): 19.88 -> 24.30Improved
move proximal anchor from client side to server side (ablation)SPECIAL · Digit-10ACC (%) and BWT (%): ACC 50.61 -> 62.12; BWT -16.42 -> -21.78Mixed
move proximal anchor from client side to server side (ablation)SPECIAL · GLUE sequence with DistilBERTACC (%): 61.37 -> 74.70Improved
increase server proximal weight lambda (ablation)SPECIAL · Digit-10, VLCS, PACS, DN4ILBWTImproved
increase server proximal weight lambda to very large values (ablation)SPECIAL · Digit-10, VLCS, PACS, DN4ILACCHurt
set lambda to 0.2 instead of selected 0.25 (ablation)SPECIAL · Digit-10ACC: -0.06%No clear effect
set lambda to 0.3 instead of selected 0.25 (ablation)SPECIAL · Digit-10ACC: -0.6%Hurt
add server-side proximal anchor to FedAvg (SPECIAL vs FedAvg)FedAvg · DN4ILACC (%): 16.47 -> 24.30Improved
add server-side proximal anchor to FedAvg (SPECIAL vs FedAvg)FedAvg · Digit-10BWT (%): -40.44 -> -21.78Improved
use SPECIAL instead of replay-based methodsSPECIAL · Digit-10, VLCS, PACS, DN4ILACC (%): 62.12 vs 61.63 (SR-FDIL) on Digit-10Improved

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