Preference Optimization for Molecular Language Models

2023/10/18 by Park, Ryan, Theisen, Ryan, Sahni, Navriti +3 · 1 citation
#Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · doi:10.48550/arxiv.2310.12304

Abstract

Molecular language modeling is an effective approach to generating novel chemical structures. However, these models do not 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.

Cited by

• Multi-domain Distribution Learning for De Novo Drug Design

Related

• Small Molecule Optimization with Large Language Models
• Grammars and reinforcement learning for molecule optimization
• Domain-Agnostic Molecular Generation with Chemical Feedback
• Large Language Models for Controllable Multi-property Multi-objective Molecule Optimization
• A Survey of Large Language Models for Text-Guided Molecular Discovery: from Molecule Generation to Optimization