Peishen Yan

FedMomentum: Preserving LoRA Training Momentum in Federated Fine-Tuning

POLAR: Policy-based Layerwise Reinforcement Learning for Stealthy Backdoor Attacks in Federated Learning