Research map / Federated learning and privacy / Personalized and heterogeneous FL
Federated Domain Generalization with Domain-specific Soft Prompts Generation
Jianhan Wu, Xiaoyang Qu, Zhangcheng Huang, Jianzong 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 |
|---|---|---|---|
| add the second-step generative prompt model (GAN) to DSPs (ablation) | FedDSPG · Office-Home | accuracy: 82.5 -> 85.3 | Improved |
| use domain-specific soft prompts instead of conditional soft prompts (ablation) | FedDSPG · Office-Home | accuracy: 84.6 -> 85.3 | Improved |
| use learnable soft prompts instead of hand-crafted prompts (ablation) | FedDSPG · Office-Home | accuracy: 77.7 -> 85.3 | Improved |
| add domain-specific prompts without generative model (ablation) | ADAPT-like (Ours-WGM) · Office-Home | accuracy: 82.9 -> 82.5 | No clear effect |
| add momentum-based aggregation (ablation) | FedDSPG · Office-Home | accuracy: 84.5 -> 85.3 | Improved |
| set momentum coefficient alpha to 0.2 instead of 1.0 (ablation) | FedDSPG · Office-Home | accuracy: 83.39 -> 85.30 | Improved |
| communicate every 2 epochs instead of 1 (ablation) | FedDSPG · Office-Home | accuracy: 85.3 -> 84.6 | Hurt |
| communicate every 0.5 epochs instead of 1 (ablation) | FedDSPG · Office-Home | accuracy: 85.3 -> 85.3 | No clear effect |
| use FedDSPG instead of FedTPG | FedTPG · DomainNet | accuracy: 72.7 -> 75.6 | Improved |
| use FedDSPG instead of FedTPG | FedTPG · DomainNet to Office-Home transfer | accuracy: 75.6 -> 79.8 | Improved |
| use FedDSPG instead of FedTPG | FedTPG · DomainNet to PACS transfer | accuracy: 90.7 -> 93.0 | Improved |
| use FedDSPG instead of ADAPT | ADAPT · DomainNet (Clp target) | accuracy: 77.5 -> 74.7 | Hurt |
Closest papers by meaning
- Harmonizing Generalization and Personalization in Federated Prompt Learning (ICML 2024 · similarity 0.91)
- DiPrompT: Disentangled Prompt Tuning for Multiple Latent Domain Generalization in Federated Learning (CVPR 2024 · similarity 0.91)
- FedPHA: Federated Prompt Learning for Heterogeneous Client Adaptation (ICML 2025 · similarity 0.90)
- Federated Learning from Vision-Language Foundation Models: Theoretical Analysis and Method (NeurIPS 2024 · similarity 0.89)
- Global Prompt Refinement with Non-Interfering Attention Masking for One-Shot Federated Learning (NeurIPS 2025 · similarity 0.89)
- Federated Domain Generalization with Data-free On-server Matching Gradient (ICLR 2025 · similarity 0.88)
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.
