SOURCE-LINKED INTELLIGENCE

Procedural Pretraining for Molecular Property Prediction

arXiv · AI, language, vision and robotics · article · Sep 15, 2026 · UTC

Molecular property prediction is often limited by the small size of labeled downstream datasets, motivating pretraining on large corpora of unlabeled molecules. In this work, we ask whether useful inductive biases can instead be learned from abstract, procedurally generated data before a model sees any molecular data. We introduce a three-stage training pipeline consisting of procedural pretraining, molecular pretraining on SMILES, and downstream fine-tuning, and evaluate several procedural tasks spanning sequence structure, cellular automata, and graph reasoning. We find that procedural pretr

Read original source ↗ Open in workspace

recordType

paper

region

Global