Research map / Federated learning and privacy / Personalized and heterogeneous FL
Rethinking Federated Prompt Learning for Medical Images: From Textual Tuning to Visual Manifold Anchoring
Yipan Wei, Wenke Huang, Yapeng Li, He Li, Qixin Zhang, Mang Ye, Bo Du
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 MSA semantic anchors to visual-prompt baseline (ablation) | FedMAP · FedISIC | accuracy: 71.55 -> 72.62 | Improved |
| add TSA topology alignment to visual-prompt baseline (ablation) | FedMAP · FedISIC | accuracy: 71.55 -> 72.80 | Improved |
| add MSA to baseline (ablation) | FedMAP · Private | accuracy: 88.00 -> 89.70 | Improved |
| add TSA to baseline (ablation) | FedMAP · Private | accuracy: 88.00 -> 88.70 | Improved |
| combine MSA and TSA (ablation) | FedMAP · FedISIC | accuracy: 72.62 -> 73.23 | Improved |
| combine MSA and TSA (ablation) | FedMAP · Private | accuracy: 89.70 -> 90.36 | Improved |
| use LLM-derived codebook ensemble instead of random anchors (ablation) | MSA · Avg of 3 datasets | accuracy: 62.67 -> 85.74 | Improved |
| use ensemble codebook instead of a single anchor per class (ablation) | MSA · Avg of 3 datasets | accuracy: 83.50 -> 85.74 | Improved |
| add Gaussian noise sigma=0.10 to anchors (ablation) | MSA · Avg of 3 datasets | accuracy: 85.74 -> 53.84 | Hurt |
| add Gaussian noise sigma=0.05 to anchors (ablation) | MSA · Avg of 3 datasets | accuracy: 85.74 -> 79.78 | Hurt |
| use MSE instead of KL divergence for relation matching (ablation) | TSA · FedISIC | accuracy: 70.99 -> 73.06 | Improved |
| use MSE instead of 1-cosine similarity (ablation) | TSA · Private | accuracy: 89.72 -> 90.36 | Improved |
Closest papers by meaning
- Global and Local Prompts Cooperation via Optimal Transport for Federated Learning (CVPR 2024 · similarity 0.87)
- FedPIA – Permuting and Integrating Adapters Leveraging Wasserstein Barycenters for Finetuning Foundation Models in Multi-Modal Federated Learning (AAAI 2025 · similarity 0.86)
- Federated Disentangled Tuning with Textual Prior Decoupling and Visual Dynamic Adaptation (ICML 2025 · similarity 0.86)
- FedVCK: Non-IID Robust and Communication-Efficient Federated Learning via Valuable Condensed Knowledge for Medical Image Analysis (AAAI 2025 · similarity 0.85)
- Unlocking the Potential of Prompt-Tuning in Bridging Generalized and Personalized Federated Learning (CVPR 2024 · similarity 0.85)
- Overcoming Heterogeneous Data in Federated Medical Vision-Language Pre-training: A Triple-Embedding Model Selector Approach (AAAI 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.
