This workshop invites high-quality submissions on federated learning (FL) for computer vision and multimodal intelligence. Topics of interest include, but are not limited to, the following areas:

(A) Foundation Models & Vision–Language FL

• Foundation-model-enhanced federated learning and knowledge distillation
• Federated transfer learning, prompt tuning, and parameter-efficient adaptation
• Optimization and scalability of large foundation models in FL

(B) Federated Learning Algorithms & Systems

• Device- and data-heterogeneous FL
• Communication- and resource-efficient optimization
• Model compression, gradient sparsification, and edge deployment
• Neural architecture search, lifelong learning, and federated domain adaptation

(C) Label-Efficient & Personalized FL

• Self-supervised, semi-supervised, and active learning in FL
• Personalized federated learning models

(D) Trustworthy & Secure FL

• Privacy leakage, privacy-preserving optimization, and machine unlearning
• Robust and secure FL against adversarial attacks
• Fairness, interpretability, and ethical considerations

(E) Applications, Datasets & Benchmarks

• FL for vision and multimodal tasks (e.g., detection, segmentation, recognition, medical imaging)
• Novel datasets, benchmarks, and evaluation protocols for federated vision
• Open-source FL frameworks and tools (e.g., FedML, Flower, OpenFL)