Extracting structured knowledge from scientific text remains a challenging task for … 展开

The majority of scientific knowledge about solid-state materials is scattered across the text, tables, and figures of millions of academic research papers. Thus, it is difficult for resea… 展开

Overall, we find excellent performance on three diverse tasks for materials science and engineering: solid-state impurity doping, metal–organic frameworks, and general material… 展开

General sequence-to-sequence NERRE
We fine-tune Llama-2 and GPT-3 models to perform NERRE tasks using 400−650 manually annotated text-extraction (prompt-completion) pair… 展开

All data used for this study are available at https://github.com/LBNLP/NERREand via Zenodo65, which contains the annotated datasets and test and train splits. Intermediate file… 展开