Research map / Federated learning and privacy / Personalized and heterogeneous FL
Beyond Description: Federated Adaptation via Semantic-Visual Prototype Alignment
Jiarong Yang, Yuan Liu
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 contrastive semantic alignment on the server instead of FedAvg prototype averaging (ablation) | FedSPA-G · 10-dataset mean, Dir(0.1) | mean accuracy: 59.18 -> 69.68 | Improved |
| add contrastive semantic alignment on the server instead of FedAvg prototype averaging (ablation) | FedSPA-P · 10-dataset mean, Dir(0.1) | mean accuracy: 64.55 -> 77.47 | Improved |
| add global semantic prototypes with alignment to local-only visual prototypes (ablation) | FedSPA-P · 10-dataset mean, Dir(0.1) | mean accuracy: 72.82 -> 77.47 | Improved |
| naive FedAvg aggregation of prototypes versus local-only adaptation (ablation) | FedAvg-based personalization · 10-dataset mean, Dir(0.1) | mean accuracy: 72.82 -> 64.55 | Hurt |
| set stability regularizer weight lambda to 0 (no regularization) (ablation) | FedSPA · 10-dataset average, RN50, 8-shot, Dir(0.1) | average accuracy | Hurt |
| use very large lambda (over-regularize toward initialization) (ablation) | FedSPA · 10-dataset average, RN50, 8-shot, Dir(0.1) | average accuracy | Hurt |
| lambda in intermediate range 0.5 to 2.0 (ablation) | FedSPA · 10-dataset average, RN50, 8-shot, Dir(0.1) | average accuracy | Improved |
| extend local training to 100 iterations in the final round | FedSPA-P · 10-dataset mean, ViT-B/16 | accuracy: 83.02 -> 84.91 | Improved |
| replace prompt tuning (pFedMoAP) with visual prototype adaptation for efficiency | FedSPA-P · efficiency, ViT-B/16 | local execution time: 83.60s -> 1.35s | Improved |
| increase shots per class from 1 to 16 | FedSPA · 10-dataset average, RN50, Dir(0.1) | accuracy | Improved |
| add feature-level Gaussian DP noise sigma=0.10 (T=5) | FedSPA-P · 10-dataset mean, RN50 | accuracy: 79.89 -> 78.80 | Hurt |
| add feature-level DP noise sigma=0.48 (T=5) | FedSPA-P · 10-dataset mean, RN50 | accuracy: 79.89 -> 72.96 | Hurt |
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.
- pFedMMA — pFedMMA: Personalized Federated Fine-Tuning with Multi-Modal Adapter for Vision-Language Models (ICLR 2026) · 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
- FedSA: A Unified Representation Learning via Semantic Anchors for Prototype-based Federated Learning (AAAI 2025 · similarity 0.88)
- Harmonizing Generalization and Personalization in Federated Prompt Learning (ICML 2024 · similarity 0.88)
- Federated Disentangled Tuning with Textual Prior Decoupling and Visual Dynamic Adaptation (ICML 2025 · similarity 0.87)
- FedPHA: Federated Prompt Learning for Heterogeneous Client Adaptation (ICML 2025 · similarity 0.87)
- BLEND: Balancing Personalization vs. Generalization in Federated Vision–Language Models (NeurIPS 2026 · similarity 0.87)
- Beyond Federated Prototype Learning: Learnable Semantic Anchors with Hyperspherical Contrast for Domain-Skewed Data (AAAI 2025 · similarity 0.87)
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.
