SAMBA: Toward an EEG Foundation Model via Differential Mamba and Coordinate-Driven Spatial Embedding

Modeling long-context electroencephalography (EEG) signals is critical for learning generalizable EEG representations, as EEG recordings are acquired at high sampling rates and often span extended durations that capture evolving neural states beyond short task-specific windows. Transformer-based models have shown promise in modeling short sequences of a few seconds; however, their quadratic complexity limits scalability to longer contexts. Moreover, substantial variability in electrode montages across datasets and recording devices poses a major challenge for building EEG models that generalize across domains. We propose SAMBA, a self-supervised learning framework with a Mamba-based U-shaped encoder-decoder architecture that effectively captures long-range temporal dependencies and montage variability in EEG data. SAMBA introduces: (1) Temporal Semantic Random Masking for semantic-level sequence reconstruction, (2) a Multi-Head Differential Mamba module to suppress redundancy and emphasize salient temporal structures, and (3) a Coordinate-Driven Spatial Embedding that learns unified embeddings in a three-dimensional Euclidean space, enabling robustness across devices. Experiments on thirteen EEG datasets across diverse tasks, electrode configurations, and sequence durations demonstrate that SAMBA achieves competitive or improved performance compared with recent state-of-the-art methods while maintaining low memory consumption and efficient inference. The learned spatial weight maps closely align with task-relevant neurophysiological regions, highlighting the interpretability of SAMBA.