Research map / Federated learning and privacy / Personalized and heterogeneous FL
Guiding Diffusion Models with Fine-Grained Conditions and Semantics-Preserving Sampling for One-Shot Federated Learning
Xiaojun Deng, Tianchi Liao, Zhiyuan Liu, Chuan Chen, Zibin Zheng
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 |
|---|---|---|---|
| add Fine-Grained Condition Learning (intra-category clusters) (ablation) | Prompts-only · DomainNet avg | accuracy: 64.00 -> 73.46 | Improved |
| add Fine-Grained Condition Learning (ablation) | Prompts-only · PACS avg | accuracy: 70.90 -> 75.30 | Improved |
| add GMM of latent noises on top of C (ablation) | Espresso · PACS avg | accuracy: 75.30 -> 79.86 | Improved |
| add Z-Sampling on top of C (ablation) | Espresso · PACS avg | accuracy: 75.30 -> 80.13 | Improved |
| add both GMM and Z-Sampling to C (ablation) | Espresso · DomainNet avg | accuracy: 73.46 -> 75.32 | Improved |
| add both GMM and Z-Sampling to C (ablation) | Espresso · Common NICO++ avg | accuracy: 80.88 -> 82.16 | Improved |
| use Espresso over FedBiP | Espresso · DomainNet quickdraw | accuracy: 70.18 -> 80.18 | Improved |
| use Espresso over FedBiP | Espresso · DomainNet sketch | accuracy: 77.74 -> 71.96 | Hurt |
| use Espresso over FedDEO | Espresso · PACS photo | accuracy: 92.05 -> 90.46 | Hurt |
| increase number of intra-category clusters J (ablation) | Espresso · DomainNet and PACS | accuracy | Improved |
| use 1 GMM component instead of 2 to 4 (ablation) | Espresso · DomainNet/PACS/NICO++ | accuracy | Hurt |
Closest papers by meaning
- One-Shot Heterogeneous Federated Learning with Local Model-Guided Diffusion Models (ICML 2025 · similarity 0.91)
- One-shot Federated Learning via Synthetic Distiller-Distillate Communication (NeurIPS 2024 · similarity 0.89)
- Exploring One-Shot Semi-supervised Federated Learning with Pre-trained Diffusion Models (AAAI 2024 · similarity 0.88)
- Revisiting Ensembling in One-Shot Federated Learning (NeurIPS 2024 · similarity 0.88)
- FuseFL: One-Shot Federated Learning through the Lens of Causality with Progressive Model Fusion (NeurIPS 2024 · similarity 0.87)
- Confusion-Resistant Federated Learning via Diffusion-Based Data Harmonization on Non-IID Data (NeurIPS 2024 · similarity 0.86)
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.
