AI Paper Atlas

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

FedBRICK: Structural Bias Aware Heterogeneous Foundation Model Federated Tuning

Yuhang Zhang, Xianda Wang, Wei Sun, Jiaxuan Chen, Fangxin Wang

AAAI 2026 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Non-IID federated 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 BRICK modules in place of missing layers to FedRAFedRA · DomainNet average (uneven layers), ViTaccuracy (%): 65.31 -> 70.33Improved
add BRICK modules to FedRAFedRA · DomainNet average (uneven layers), Mixeraccuracy (%): 52.71 -> 56.28Improved
add BRICK modules to FedRAFedRA · NICO++ average (uneven layers), ViTaccuracy (%): 86.79 -> 88.11Improved
add BRICK modules to FedRA on quickdraw domain with 6 layersFedRA · DomainNet quickdraw (6 layers), ViTaccuracy (%): 26.94 -> 46.98Improved
add BRICK modules when every client keeps only 4 layersFedRA · DomainNet average (all clients 4 layers), ViTaccuracy (%): 14.90 -> 43.86Improved
add BRICK modules when every client keeps only 4 layersFedRA · DomainNet average (all clients 4 layers), Mixeraccuracy (%): 14.52 -> 17.18Improved
add BRICK modules when every client keeps 6 layersFedRA · DomainNet average (all clients 6 layers), ViTaccuracy (%): 53.45 -> 65.77Improved
plug BRICK into DepthFL and InclusiveFL as wellDepthFL, InclusiveFL, FedRA · DomainNet and NICO++ (ViT)accuracy (%)Improved
add BRICK modules to FedRAFedRA · DomainNet client representationscosine distance across clientsImproved
gain from BRICK is larger on clients with fewer layersFedBRICK · DomainNet and NICO++accuracy (%): +12.32 (6 layers) vs +28.97 (4 layers), ViTImproved
add BRICK to Mixer when all clients have 4 layers (clipart)FedRA · DomainNet clipart (4 layers), Mixeraccuracy (%): 20.33 -> 20.50No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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