CVPR 2026 accepted papers on Federated learning and privacy
50 of the 4,042 papers accepted at CVPR 2026 (1.2%) are on Federated learning and privacy, compared with 0.9% at CVPR 2025. They fall into 4 topics.
See CVPR 2026 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Personalized and heterogeneous FL · 29 papers
- Domain-Skewed Federated Learning with Feature Decoupling and Calibration
- Fed-ADE: Adaptive Learning Rate for Federated Post-adaptation under Distribution Shift
- FedAFD: Multimodal Federated Learning via Adversarial Fusion and Distillation
- FedBPrompt: Federated Domain Generalization Person Re-Identification via Body Distribution Aware Visual Prompts
- FedDAP: Domain-Aware Prototype Learning for Federated Learning under Domain Shift
- Federated Active Learning Under Extreme Non-IID and Global Class Imbalance
- FedHarmony: Harmonizing Heterogeneous Label Correlations in Federated Multi-Label Learning
- FedMPT: Federated Multi-Label Prompt Tuning of Vision-Language Models
- FedRE: A Representation Entanglement Framework for Model-Heterogeneous Federated Learning
- FedRG: Unleashing the Representation Geometry for Federated Learning with Noisy Clients
- Few-for-Many Personalized Federated Learning
- Fine-Tuning Impairs the Balancedness of Foundation Models in Long-tailed Personalized Federated Learning
and 17 more in the interactive map.
Differential privacy · 10 papers
- Differentially Private 2D Human Pose Estimation
- DP-FedAdamW: An Efficient Optimizer for Differentially Private Federated Large Models
- Fully Decentralized Certified Unlearning
- LDP-Slicing: Local Differential Privacy for Images via Randomized Bit-Plane Slicing
- Personalized Federated Training of Diffusion Models with Privacy Guarantees
- Reinforcement-Guided Synthetic Data Generation for Privacy-Sensitive Identity Recognition
- Taming Noise-Induced Prototype Degradation for Privacy-Preserving Personalized Federated Fine-Tuning
- Unsafe2Safe: Controllable Image Anonymization for Downstream Utility
- FedAlign: Differentially Private Distribution Alignment for Non-IID Federated Learning
- PrivSynth: Alternating and Control-Based Optimization for Privacy and Utility in Synthetic Data
Privacy attacks and secure learning · 7 papers
- Protego: User-Centric Pose-Invariant Privacy Protection Against Face Recognition-Induced Digital Footprint Exposure (1 citation)
- Revisiting Geometric Obfuscation with Dual Convergent Lines for Privacy-Preserving Image Queries in Visual Localization
- Batman: Benign Knowledge Alignment Through Malicious Null Space in Federated Backdoor Attack
- Eliminate Distance Differences Induced by Backdoor Attacks: Layer-Selective Training and Clipping to Mask Backdoor Models
- FedMOP: Achieving Enhanced Privacy and Performance in Federated Learning via Momentum Orthogonal Projection
- PrivateEyes: Gaze-Preserving Anonymization for Data Sharing
- Stealing Split Learning Bottom Models by Recovering Embedding Geometry