GRAFT: Gain-Recalibrated Adapters for Transformer-Based Neural Population Activity Modeling
Researchers have developed GRAFT, a Transformer-based model designed for neural population activity modeling. This new model separates reusable temporal dynamics from a recalibratable neuron interface, allowing for better adaptation in brain-computer interfaces where neuron identities and statistics can change. GRAFT achieved a new state-of-the-art performance on the NLB'21 protocol, reaching 0.3866 co-bps. Furthermore, it demonstrated efficient cross-day recalibration by updating only a small percentage of its parameters. AI
IMPACT Sets new SOTA on neural population activity modeling, potentially improving brain-computer interfaces.