Augmenting Molecular Language Models with Local $n$-gram Memory

arXiv:2606.12113v1 Announce Type: cross Abstract: Transformer-based language models for SMILES strings suffer from a locality gap: standard character-level tokenization fragments chemically meaningful motifs, forcing models to repeatedly learn local syntax at the expense of long-range dependencies. To address this without disrupting standard...

Read the full article at the source.