New methods on using Large Language Models to Extract Materials Databases from Literature from Skunkworks Team

Image created by Maciej Polak, lead author of the papers below, using machine learning tools from stabilityAI

We are excited to share two new papers showing great capabilities for large language models to extract data from papers. This work originated as a Skunkworks project. Congratulations to undergraduates Shrey Modi, Anna Latosinska, Jinming Zhang, Ching-Wen Wang, Shanonan Wang, and Ayan Deep Hazra for their contributions and coauthorship on the first paper.

News piece from UW CoE:

Paper 1 – Flexible, Model-Agnostic Method for Materials Data Extraction from Text Using General Purpose Language Models –

Paper 2 – Extracting Accurate Materials Data from Research Papers with Conversational Language Models and Prompt Engineering — Example of ChatGPT –