This optimizes the TEE Area utilization and improves the model protection and efficiency, considerably reducing storage requirements, particularly in useful resource-constrained federated learning scenarios. In the https://aistudio.ml