1 citations · 1 across the 4 of their papers we have counts for
1 paper · 2 filters
Yuting Ma, Lechao Cheng, Xiaohua Xu
Federated Learning (FL) with pre-trained Vision-Language Models (VLMs) has emerged as a promising paradigm for various downstream tasks. By leveraging its strong representations, r…