Accelerated materials language processing enabled by GPT

TL;DR

This paper demonstrates that GPT-based prompt engineering can effectively replace complex traditional models for materials language processing tasks, achieving comparable or improved results with less data and effort.

Contribution

The study introduces GPT-enabled pipelines for document classification, NER, and extractive QA in materials science, simplifying architecture and reducing data requirements.

Findings

GPT-based classification matches prior accuracy with small datasets

Few-shot prompts improve entity recognition performance

GPT-enabled QA can automatically correct annotations

Abstract

Materials language processing (MLP) is one of the key facilitators of materials science research, as it enables the extraction of structured information from massive materials science literature. Prior works suggested high-performance MLP models for text classification, named entity recognition (NER), and extractive question answering (QA), which require complex model architecture, exhaustive fine-tuning and a large number of human-labelled datasets. In this study, we develop generative pretrained transformer (GPT)-enabled pipelines where the complex architectures of prior MLP models are replaced with strategic designs of prompt engineering. First, we develop a GPT-enabled document classification method for screening relevant documents, achieving comparable accuracy and reliability compared to prior models, with only small dataset. Secondly, for NER task, we design an entity-centric prompts, and learning few-shot of them improved the performance on most of entities in three open datasets. Finally, we develop an GPT-enabled extractive QA model, which provides improved performance and shows the possibility of automatically correcting annotations. While our findings confirm the potential of GPT-enabled MLP models as well as their value in terms of reliability and practicability, our scientific methods and systematic approach are applicable to any materials science domain to accelerate the information extraction of scientific literature.