AI Paper Atlas

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

Federated Vision-Language-Recommendation with Personalized Fusion

Zhiwei Li, Guodong Long, Jing Jiang, Chengqi Zhang, Qiang Yang

AAAI 2026 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Federated prompt and LoRA tuning · 13 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 FedVLR personalized fusionFedRAP · BeautyHR@50: 23.72 -> 38.34Improved
add FedVLR personalized fusionFedAvg · KUHR@50: 11.27 -> 54.90Improved
add FedVLR personalized fusionFedNCF · KUHR@50: 11.76 -> 43.63Improved
add FedVLR personalized fusionFedNCF · MLHR@50: 3.93 -> 11.31Improved
replace BLFM with generic MLP fusion (ablation)FedVLR · BeautyHR@50: FCF: 37.15 -> 41.11Mixed
replace BLFM with generic MLP fusion (ablation)FedVLR on FedRAP · CardsHR@50: 73.74 -> 64.19Hurt
naive federated VBPR instead of BLFM (ablation)FedAvg · larger datasetsHR@50Hurt
remove ID, text or vision features (ablation)FedVLR · KUHR@50Improved
add Gaussian gradient noiseFedVLR · KU, FedNCFHR@50: 19.61 -> 16.80Hurt
add Gaussian gradient noiseFedVLR · KU, FedRAPHR@50: 38.24 -> 37.25Hurt
use FedVLR instead of centralized modelsFedRAP + FedVLR · BeautyHR@50: 29.25 (MGCN) -> 38.34Improved
use FedVLR instead of centralized modelsFedRAP + FedVLR · FoodHR@50: 36.16 (PGL) -> 11.42Hurt

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