Publication
Empowering Molecule Discovery for Molecule-Caption Translation with Large Language Models: A ChatGPT Perspective
Molecule Discovery, Large Language Models
Cite this work View paper
Citation
Jiatong Li, Yunqing Liu, Wenqi Fan, Xiao-yong Wei, Hui Liu, Jiliang Tang, and Qing Li. (2024). Empowering Molecule Discovery for Molecule-Caption Translation with Large Language Models: A ChatGPT Perspective. IEEE TKDE.
About this work
MolReGPT uses retrieved examples to translate between molecules and natural-language descriptions. It selects similar molecules and their captions from a local database, then includes them in a language model’s prompt.
The method uses in-context learning without fine-tuning. Experiments evaluate molecule understanding and text-based molecule generation, including comparisons with MolT5-base.
Our codes are available here: Code