SOURCE-LINKED INTELLIGENCE

Self-Supervised Pretraining of Molecular Graph Encoders with LeJEPA

arXiv · AI, language, vision and robotics · article · Sep 2, 2026 · UTC

Self-supervised pretraining has transformed language and vision, but its value for molecular graph neural networks remains contested. We ask whether pretraining on a large unlabelled corpus improves molecular property prediction. We adapt LeJEPA, a predictor-free joint-embedding predictive architecture regularised by Sketched Isotropic Gaussian Regularisation (SIGReg), to molecular graphs, evaluating GPS and Chemprop-style D-MPNN encoders on the Wong et al. [1] antibiotic-activity dataset and ogbg-molhiv using a multi-seed, bootstrap-based protocol. Pretraining improves learned representations

Read original source ↗ Open in workspace

recordType

paper

region

Global