Preference Optimization for Molecular Language Models
This addresses the need for better control over generated molecules in chemistry, but it is incremental as it applies an existing method to a new domain.
The paper tackled the problem of aligning molecular language models with chemist preferences by fine-tuning using Direct Preference Optimization, resulting in a simple, efficient, and highly effective approach.
Molecular language modeling is an effective approach to generating novel chemical structures. However, these models do not \emph{a priori} encode certain preferences a chemist may desire. We investigate the use of fine-tuning using Direct Preference Optimization to better align generated molecules with chemist preferences. Our findings suggest that this approach is simple, efficient, and highly effective.