Search papers, labs, and topics across Lattice.
3
0
3
A single EEG decoding architecture, DSAINet, achieves state-of-the-art generalizability across diverse tasks and datasets without task-specific tuning, despite having only 77K parameters.
Forget masked reconstruction: diffusion-based generative pretraining unlocks stronger fMRI foundation models that generalize across diverse brain states and tasks.
By fusing temporal and spatial EEG features layer-by-layer, LI-DSN shatters the "information silo" problem of traditional dual-stream networks, unlocking significant gains in decoding accuracy across diverse BCI tasks.
