Preprint
Aug 2026
Geometry-Informed Parameter-Efficient Fine-Tuning of Pre-trained Molecular GNNs for Blood-Brain Barrier Permeability Prediction
BBBP-GeoPEFT is proposed, a geometry-informed PEFT framework for pre-trained molecular GNNs that preserves pre-trained knowledge while incorporating permeability-relevant geometric information with a small trainable-parameter budget.
Marco Vieto Vega, Long D. Nguyen, Binh P. Nguyen
· 0 citations