Self-supervised foundation models have recently shown strong potential for electroencephalogram (EEG)-based analysis. However, existing approaches struggle to capture the inherently multi-scale temporal structure of EEG signals, where local neural patterns and long-range dependen…
ABSTRACT With the rapid development of intelligent technologies, human‐machine interfaces (HMIs) are increasingly integrated into modern society, enabling efficient communication between humans and robots and thus building intelligent living. Flexible sensors are key signal conve…