CVPR 2025 accepted papers on Federated learning and privacy

Personalized and heterogeneous FL · 16 papers

• FedAWA: Adaptive Optimization of Aggregation Weights in Federated Learning Using Client Vectors (28 citations)
• Patient-Level Anatomy Meets Scanning-Level Physics: Personalized Federated Low-Dose CT Denoising Empowered by Large Language Model (18 citations)
• FedBiP: Heterogeneous One-Shot Federated Learning with Personalized Latent Diffusion Models (14 citations)
• A Simple Data Augmentation for Feature Distribution Skewed Federated Learning (13 citations)
• Federated Learning with Domain Shift Eraser (12 citations)
• Beyond Local Sharpness: Communication-Efficient Global Sharpness-aware Minimization for Federated Learning (8 citations)
• Geometric Knowledge-Guided Localized Global Distribution Alignment for Federated Learning (8 citations)
• FedCALM: Conflict-aware Layer-wise Mitigation for Selective Aggregation in Deeper Personalized Federated Learning (7 citations)
• HistoFS: Non-IID Histopathologic Whole Slide Image Classification via Federated Style Transfer with RoI-Preserving (7 citations)
• FedSPA: Generalizable Federated Graph Learning under Homophily Heterogeneity (6 citations)
• dFLMoE: Decentralized Federated Learning via Mixture of Experts for Medical Data Analysis (5 citations)
• F^3OCUS - Federated Finetuning of Vision-Language Foundation Models with Optimal Client Layer Updating Strategy via Multi-objective Meta-Heuristics (5 citations)

and 4 more in the interactive map.

Privacy attacks and secure learning · 9 papers

• Model Poisoning Attacks to Federated Learning via Multi-Round Consistency (28 citations)
• Unlearning through Knowledge Overwriting: Reversible Federated Unlearning via Selective Sparse Adapter (22 citations)
• FedMIA: An Effective Membership Inference Attack Exploiting "All for One" Principle in Federated Learning (21 citations)
• Detecting Backdoor Attacks in Federated Learning via Direction Alignment Inspection (20 citations)
• NoT: Federated Unlearning via Weight Negation (12 citations)
• Fortifying Federated Learning Towards Trustworthiness via Auditable Data Valuation and Verifiable Client Contribution (7 citations)
• From Laboratory to Real World: A New Benchmark Towards Privacy-Preserved Visible-Infrared Person Re-Identification (7 citations)
• Gradient Inversion Attacks on Parameter-Efficient Fine-Tuning (6 citations)
• Split Adaptation for Pre-trained Vision Transformers (2 citations)

Communication-efficient FL · 1 paper

• AFL: A Single-Round Analytic Approach for Federated Learning with Pre-trained Models (6 citations)