ST-CoG-XAI: A Spectrotemporal Contrastive Generation Foundation Model for Explainable EEG Decoding

Wearable electroencephalography (EEG) on flexible substrates enables continuous neuromonitoring but violates the fixed-montage assumption of conventional decoders: electrode geometry varies across devices, channels intermittently disappear, and task-relevant biomarkers drift over spectral and spatial dimensions. We present ST-CoG-XAI, a unified framework that delivers accurate and intrinsically interpretable decoding under these nonstationary conditions. The model begins with dual-stream encoder: an asymmetric convolutional network converts time-domain windows into patch tokens, whereas a graph neural network tokenizes frequency-band connectomes. These two token streams are fused and processed by a Unified Spectral-Spatial-Temporal Synergy Criss-Cross Transformer (UniSST-CCT) whose multi-head attention is decomposed into spatial, temporal, and spectral groups, capturing cross-axis dependencies at only a constant-factor increase in complexity. Prior to supervision, the encoder is exposed to large unlabelled data through Spectro-Temporal Contrastive Generation (CoG), which alternates three mutually exclusive reconstruction tasks to align temporal and spectral representations. After CoG pre-training, supervised fine-tuning activates a weight-sharing second forward pass whose DeepLIFT-guided re-attention (XAI) refines the encoder while preserving clinically meaningful patterns. Evaluated on seven clinical BCI benchmarks, covering neurodegenerative disorders, epilepsy detection, sleep staging, and event-related classification, ST-CoG-XAI matches or surpasses published state-of-the-art results, confirming robust generalization not only across heterogeneous EEG montages but also when operating with a limited or partial set of channels.