AI Paper Atlas

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

Improving Model Fusion by Training-Time Neuron Alignment With Fixed Neuron Anchors

Zexi Li, Zhiqi Li, Jie Lin, Tao Shen, Jun Xiao, Yike Guo, Tao Lin, Chao Wu

TPAMI 2025 · 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
add TNA-PFNvanilla training · CIFAR-10 CNN 5-model fusioninterpolated accuracy: 21.3 -> 48.3Improved
add TNA-PFNvanilla training · MNIST MLP_h5_w200 5-model fusionaccuracy barrier: 0.643 -> 0.0974Improved
add TNA-PFNpretrained ResNet18 · CIFAR-10 LMCinterpolated accuracy: 11.9 -> 54.9Improved
add TNA-PFNsemi-trained CNN · CIFAR-10 LMCaccuracy barrier: 0.181 -> 0.0413Improved
add TNA-PFNgreedy soup of CLIP ViT-B/32 · CIFAR-100 different seedsaccuracy: 67.7 -> 73.3Improved
add TNA-PFNuniform soup of CLIP ViT-B/32 · CIFAR-10 different learning ratesaccuracy: 91.6 -> 91.2Hurt
add TNA-PFNmixed model soups · CIFAR-10 / CIFAR-100accuracy: CIFAR-100 greedy 80.9 -> 82.6; CIFAR-10 uniform 91.6 -> 91.2Mixed
add TNA-PFN to FedAvg (ColD fusion)FedAvg · GLUE COLA+RTE+MRPC Sumaccuracy: 77.66 -> 81.43Improved
use FedPFNFedAvg · CIFAR-10 CNN dir 0.1top-1 accuracy: 57.5 -> 62.2Improved
use FedPNUFedAvg · CIFAR-100 ResNet20 dir 0.1top-1 accuracy: 28.5 -> 35.3Improved
add FedPFN (resample mask each round) instead of fixing the mask (ablation)FedPFN · CIFAR-10 CNN dir 0.3top-1 accuracy: 64.9 -> 65.7Improved
use gradient-mask FedPFN instead of random pruning at init (ablation)FedPruning · CIFAR-10 CNN dir 0.3top-1 accuracy: 63.7 -> 65.7Improved

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