Chem-GMNet uses sphere-native embeddings, DualSKA attention, and SH-FFN layers to match or beat ChemBERTa-2 on MoleculeNet tasks with fewer parameters and sometimes no pretraining.
Michael Poli, Stefano Massaroli, Eric Nguyen, et al
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 3roles
background 1polarities
background 1representative citing papers
TIGER is a text-informed ML framework that improves bidirectional enzyme-reaction retrieval by distilling semantic knowledge from sequences and aligning representations across tasks and distributions.
A systematic survey and benchmark of four deep learning paradigms for molecular property prediction that organizes the field, critiques current data practices, and outlines three future directions.
citing papers explorer
-
Chem-GMNet: A Sphere-Native Geometric Transformer for Molecular Property Prediction
Chem-GMNet uses sphere-native embeddings, DualSKA attention, and SH-FFN layers to match or beat ChemBERTa-2 on MoleculeNet tasks with fewer parameters and sometimes no pretraining.
-
TIGER: Text-Informed Generalized Enzyme-Reaction Retrieval
TIGER is a text-informed ML framework that improves bidirectional enzyme-reaction retrieval by distilling semantic knowledge from sequences and aligning representations across tasks and distributions.