BVFLMSP : Bayesian Vertical Federated Learning for Multimodal Survival with Privacy

ArXi:2604.02248v1 Announce Type: cross Multimodal time-to-event prediction often requires integrating sensitive data distributed across multiple parties, making centralized model