Research map / Federated learning and privacy / Communication-efficient FL
EFTViT: Efficient Federated Training of Vision Transformers with Masked Images on Resource-Constrained Clients
Meihan Wu, Tao Chang, Cui Miao, Jie Zhou, Chun Li, Xiangyu Xu, Ming Li, Xiaodong Wang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| increase masking ratio from 0.00 to 0.75 (ablation) | EFTViT · client training cost | GFLOPs: 12.005 -> 2.997 | Improved |
| increase masking ratio to 0.95 (beyond 0.75) (ablation) | EFTViT · CIFAR-100 | accuracy | Hurt |
| increase masking ratio from 0 to 75% (ablation) | EFTViT · CIFAR-100 | accuracy: accuracy remains larger than 90% | No clear effect |
| use 4 local layers instead of 2 (ablation) | EFTViT · CIFAR-100 | accuracy: 90.81 -> 90.14 | Hurt |
| use 6 local layers instead of 2 (ablation) | EFTViT · CIFAR-100 | accuracy: 90.81 -> 89.37 | Hurt |
| use 6 local layers instead of 2 (ablation) | EFTViT · UC Merced Land-Use | accuracy: 98.80 -> 97.38 | Hurt |
| use 4 local layers instead of 2 (ablation) | EFTViT · CIFAR-10 | accuracy: 98.35 -> 97.56 | Hurt |
| raise median sampling threshold higher (ablation) | EFTViT · CIFAR-10/100, UC Merced | accuracy | No clear effect |
| hierarchical masked training vs FedRA | EFTViT · UC Merced Land-Use (beta=0.1) | accuracy: 96.43 -> 98.80 | Improved |
| hierarchical masked training vs FedRA | EFTViT · UC Merced Land-Use | rounds to 85% accuracy: 55 -> 6 | Improved |
| hierarchical masked training vs FedRA | EFTViT · UC Merced Land-Use | training time per round (s): 6.436 -> 2.025 | Improved |
| hierarchical masked training vs Fed-LoRA | EFTViT · CIFAR-100 (beta=0.1) | accuracy: 70.23 -> 90.02 | Improved |
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.
- FedRA — FedRA: A Random Allocation Strategy for Federated Tuning to Unleash the Power of Heterogeneous Clients (ECCV 2024) · better on 13 of 13 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
- Closest Neighbors are Harmful for Lightweight Masked Auto-encoders (CVPR 2025 · similarity 0.86)
- Unlocking the Potential of Prompt-Tuning in Bridging Generalized and Personalized Federated Learning (CVPR 2024 · similarity 0.86)
- Federated Learning via Input-Output Collaborative Distillation (AAAI 2024 · similarity 0.85)
- FedDAT: An Approach for Foundation Model Finetuning in Multi-Modal Heterogeneous Federated Learning (AAAI 2024 · similarity 0.85)
- PRISM: Privacy-Preserving Improved Stochastic Masking for Federated Generative Models (ICLR 2025 · similarity 0.84)
- F^3OCUS - Federated Finetuning of Vision-Language Foundation Models with Optimal Client Layer Updating Strategy via Multi-objective Meta-Heuristics (CVPR 2025 · similarity 0.84)
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.
