Preference Optimization for Molecular Language Models
Abstract
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.
Cite
Text
Park et al. "Preference Optimization for Molecular Language Models." NeurIPS 2023 Workshops: GenBio, 2023.Markdown
[Park et al. "Preference Optimization for Molecular Language Models." NeurIPS 2023 Workshops: GenBio, 2023.](https://mlanthology.org/neuripsw/2023/park2023neuripsw-preference/)BibTeX
@inproceedings{park2023neuripsw-preference,
title = {{Preference Optimization for Molecular Language Models}},
author = {Park, Ryan and Theisen, Ryan and Rahman, Rayees and Cichońska, Anna and Patek, Marcel and Sahni, Navriti},
booktitle = {NeurIPS 2023 Workshops: GenBio},
year = {2023},
url = {https://mlanthology.org/neuripsw/2023/park2023neuripsw-preference/}
}