L+M-24: Building a Dataset for Language + Molecules @ ACL 2024
arXiv:2403.00791
Abstract
Language-molecule models have emerged as an exciting direction for molecular discovery and understanding. However, training these models is challenging due to the scarcity of molecule-language pair datasets. At this point, datasets have been released which are 1) small and scraped from existing databases, 2) large but noisy and constructed by performing entity linking on the scientific literature, and 3) built by converting property prediction datasets to natural language using templates. In this document, we detail the dataset, which has been created for the Language + Molecules Workshop shared task at ACL 2024. In particular, is designed to focus on three key benefits of natural language in molecule design: compositionality, functionality, and abstraction.
The dataset, finetuned baselines, and evaluation code are released publicly at https://github.com/language-plus-molecules/LPM-24-Dataset through https://huggingface.co/language-plus-molecules