AI Paper Atlas

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

Decoupled Training with Local Reinforcement Fine-Tuning in Federated Learning

Yuting Ma, Lechao Cheng, Xiaohua Xu

ICML 2026 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Federated prompt and LoRA tuning · 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 decoupled encoder training (server text encoder, client image encoders) (ablation)FedLoRA · Full Non-IID, seven datasetsbase accuracy: 58.11 -> 86.68Improved
add decoupled encoder training (ablation)FedLoRA · Few Non-IID, seven datasetsbase accuracy: 78.32 -> 86.42Improved
add decoupled encoder training (ablation)FedLoRA · IID settings, seven datasetsnovel accuracy: 77.44 -> 75.35 (Few IID)Hurt
add two-stage local fine-tuning (SFT then RL) (ablation)FedLoRA · Few IID, seven datasetsnovel accuracy: 77.44 -> 83.11Improved
add two-stage local fine-tuning alone (ablation)FedLoRA · Full Non-IID, seven datasetsbase accuracy: 58.11 -> 47.91Hurt
combine decoupled encoder and two-stage fine-tuning (ablation)FedLoRA · Full Non-IID, seven datasetsharmonic mean: 63.12 -> 85.03Improved
use RL-only local fine-tuning instead of SFT-only (ablation)local fine-tuning · base-to-novel datasets under communication roundsnovel accuracyImproved
use SFT-only local fine-tuning (ablation)local fine-tuning · base-to-novel datasets under communication roundsbase and novel accuracyHurt
use mixed reference model instead of latest or final-SFT only (ablation)RL stage · few-shot vs full-dataaccuracyMixed
vary image embedding upload ratio from 0.2 to 1.0 (ablation)FedDTL · Full IID and Full Non-IIDaccuracyNo clear effect
add noise perturbation to uploaded embeddings (ablation)FedDTL · Full Non-IID, nine datasetsharmonic mean: 83.12 -> 84.15Improved
add embedding aggregation to uploaded embeddings (ablation)FedDTL · Full Non-IID, nine datasetsharmonic mean: 83.12 -> 76.65Hurt

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