1 paper
Juan Zhu, Zixin Wang, Shenghui Song +2
Foundation models (FMs) unlock unprecedented multimodal and multitask intelligence, yet their cloud-centric deployment precludes real-time responsiveness and compromises user priva…