Research map / Federated learning and privacy / Communication-efficient FL
OS-Fed: One Snapshot Is All You Need
Xuwei Qian, Jinghui Zhang, Yuchuan Tan, Wenbo Huang, Zhen Wu, Shen Zhou, LiSha Gao, Ding Ding, Fang Dong
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 |
|---|---|---|---|
| use OS-FED single snapshot instead of full FedAvg updates | FedAvg (RegNetX-4.0GF) · ImageNet-10 | total traffic (MB): 8040 MB -> 57 MB, accuracy 0.65 -> 0.64 | Improved |
| use multi-grid end-to-end snapshot instead of optimize-then-downsample (ablation) | OS-FED · ImageNet-10 (ResNet-18) | accuracy: 48.8 -> 63.5 | Improved |
| add normalization to the L2 snapshot-matching loss (ablation) | OS-FED · ImageNet-10 (ResNet-18) | accuracy: 61.0 -> 63.5 | Improved |
| use single-level optimization instead of bi-level optimization (ablation) | OS-FED · ImageNet-10 (ResNet-18) | accuracy: 28.3 -> 61.0 | Improved |
| use L2 distance instead of cosine similarity (ablation) | OS-FED · ImageNet-10 (ResNet-18) | accuracy: 57.4 -> 61.0 | Improved |
| add error compensation (ablation) | OS-FED · ImageNet-10 | accuracy | Improved |
| use multi-grid snapshot instead of single monolithic snapshot (ablation) | OS-FED · ImageNet-10 | snapshot quality | Improved |
| increase grid size from 2x2 to 4x4 (ablation) | OS-FED · CIFAR-100 | accuracy: 26.9% -> 30.3% | Improved |
| add noise during snapshot optimization for privacy (ablation) | OS-FED · ImageNet-1K (60 clients) | Top-1 accuracy: 35.1% -> 33.5%, CosSim 0.122 -> 0.05 | Hurt |
| OS-FED versus DP-SGD for privacy | OS-FED · ImageNet-1K (60 clients) | wall-time per client: 10.1s -> 3.8s | Improved |
| OS-FED versus RandTopk, LoRA-FAIR on CIFAR-100 | LoRA-FAIR · CIFAR-100 | accuracy: 26.3% -> 30.3% | Improved |
| OS-FED on FMNIST with MLP versus FedAF | FedAF · FMNIST (MLP, 10 clients) | accuracy: 84.1% -> 83.9% | No 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.
- LoRA-FAIR — LoRA-FAIR: Federated LoRA Fine-Tuning with Aggregation and Initialization Refinement (ICCV 2025) · better on 3 of 3 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
- FedBiP: Heterogeneous One-Shot Federated Learning with Personalized Latent Diffusion Models (CVPR 2025 · similarity 0.87)
- An Efficient and Accurate Dynamic Sparse Training Framework Based on Parameter-Freezing (AAAI 2025 · similarity 0.86)
- One-shot Federated Learning via Synthetic Distiller-Distillate Communication (NeurIPS 2024 · similarity 0.86)
- Little Is Enough: Boosting Privacy by Sharing Only Hard Labels in Federated Semi-Supervised Learning (AAAI 2025 · similarity 0.85)
- Revisiting Ensembling in One-Shot Federated Learning (NeurIPS 2024 · similarity 0.85)
- Achieving Lossless Gradient Sparsification via Mapping to Alternative Space in Federated Learning (ICML 2024 · similarity 0.85)
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.
