Abstract: In the task of automatic sleep stage classification, deep learning models often face the challenge of balancing temporal-spatial feature extraction with computational complexity. To address ...