Zezhou Wang

federated learning (1) reinforcement learning (1) vision transformer (1) performance optimization (1) domain adaptation (1) data augmentation (1) code generation (1) instruction tuning (1) fisher information (1) off-policy learning (1) language model (1) vision-language model (1) proximal policy optimization (1) computational cost (1) token pruning (1) federated instruction tuning (1) extended berkeley packet filter (1) kernel bypass (1) state machine replication (1) distributed protocol (1)