SOURCE-LINKED INTELLIGENCE

Structural Hierarchy and Geometry in Molecular Representation Learning

arXiv · AI, language, vision and robotics · article · Aug 30, 2026 · UTC

Molecular self-supervised learning uses chemical structures to guide which molecular embeddings should be similar. We study whether explicitly encoding a molecule's Bemis-Murcko scaffold and using it to supervise the molecular embedding changes what the model learns. We further test whether this effect depends on the embedding geometry by comparing Euclidean and Lorentz contrastive objectives. Across two augmentation strengths, scaffold-supervised models consistently organize molecules according to both identical and structurally related scaffolds. The resulting embeddings also improve molecul

Read original source ↗ Open in workspace

recordType

paper

region

Global