Daguang Xu

Latest

• FedBPT: Efficient Federated Black-box Prompt Tuning for Large Language Models
• GradViT: Gradient Inversion of Vision Transformers
• Do Gradient Inversion Attacks Make Federated Learning Unsafe?