AI Paper Atlas

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

pFedMMA: Personalized Federated Fine-Tuning with Multi-Modal Adapter for Vision-Language Models

Sajjad Ghiasvand, Mahnoosh Alizadeh, Ramtin Pedarsani

ICLR 2026 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Federated prompt and LoRA tuning · 9 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
use pFedMMA instead of FedPGPpFedMMA · CLIP datasets 16-shot (avg 7)HM accuracy (%): 79.09 -> 84.15Improved
use pFedMMA instead of pFedMoAP on local classespFedMMA · CLIP datasets 16-shot (avg 7)local accuracy (%): 97.89 -> 97.17Hurt
use pFedMMA on DomainNet instead of FedPGPpFedMMA · DomainNet (Dirichlet 0.5)avg accuracy (%): 24.90 -> 47.17Improved
use pFedMMA on Office-Caltech10 instead of FedPGPpFedMMA · Office-Caltech10 (Dirichlet 0.5)avg accuracy (%): 20.71 -> 21.33Improved
use pFedMMA instead of pFedMoAP at high label skewpFedMMA · CIFAR-100 (beta=0.1)accuracy (%): 80.29 -> 81.82Improved
use pFedMMA at mild heterogeneity (beta=10) vs FedPGPpFedMMA · CIFAR-100 (beta=10)accuracy (%): 73.90 -> 73.65No clear effect
keep up/down projections local, aggregate shared projection (vs all global) (ablation)pFedMMA · DTDaccuracy (%): 61.10 -> 76.38Improved
keep up/down local and aggregate shared projection (vs shared local, up/down global) (ablation)pFedMMA · DTDaccuracy (%): 62.19 -> 76.38Improved
use a single multi-modal shared projection instead of vision-only aggregation (ablation)pFedMMA · CLIP datasetsHM accuracy (%): 79.31 -> 79.70Improved
raise scaling factor alpha from 0.0001 to 0.005 (ablation)pFedMMA · CLIP datasetsHM accuracy (%): 7.62 -> 83.25Improved
raise scaling factor alpha from 0.005 to 0.01 (ablation)pFedMMA · CLIP datasetsHM accuracy (%): 83.25 -> 82.03Hurt
insert adapters in last three layers instead of last layer only (ablation)pFedMMA · CLIP datasetsHM accuracy (%): 83.57 -> 84.38Improved

Later papers that compared with this method

2 later accepted papers report pFedMMA as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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