English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

This study is dedicated to evaluating the capabilities of advanced largelanguage models (LLMs) such as GPT-3.5-Turbo, GPT-4, and GPT-4-Turbo in theextraction of structured information from scientific documents within the fieldof materials science. We introduce a novel methodology for the comparativeanalysis of intricate material expressions, emphasising the standardisation ofchemical formulas to tackle the complexities inherent in materials scienceinformation assessment. To this end, we primarily focus on two critical tasksof information extraction: (i) a named entity recognition (NER) of studiedmaterials and physical properties and (ii) a relation extraction (RE) betweenthese entities. The performance of LLMs in executing these tasks is benchmarkedagainst traditional models based on the BERT architecture and rule-basedapproaches. For NER, LLMs fail to outperform the baseline with zero-shotprompting and exhibit only limited improvement with few-shot prompting.However, for RE, a GPT-3.5-Turbo fine-tuned with the appropriate strategyoutperforms all models, including the baseline. Without any fine-tuning, GPT-4and GPT-4-Turbo display remarkable reasoning and relationship extractioncapabilities after being provided with merely a couple of examples, surpassingthe baseline. Overall, the results suggest that although LLMs demonstraterelevant reasoning skills in connecting concepts, for tasks requiringextracting complex domain-specific entities like materials, specialised modelsare currently a better choice.

Mining experimental data from Materials Science literature with Large Language Models