Scholarlink

<p>Scientific text is a promising source of data in materials science, with ongoing research into utilising textual data for materials discovery. In this study, we developed and tested a Question Answering (QA) approach to extract material-property relationships from scientific publications. QA performance was evaluated for information extraction of perovskite bandgaps based on a human query. We observed considerable variation in results with five different large language models fine-tuned for the QA task. Best extraction accuracy was achieved with the QA MatSciBERT and F1-scores improved on the current state-of-the-art. QA also outperformed three latest generative large language models on the information extraction task, except the GPT-4 model. This work demonstrates the QA workflow and paves the way towards further applications. The simplicity and versatility of the QA approach all point to its considerable potential for text-driven discoveries in materials research.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Question Answering models for information extraction from perovskite materials science literature

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月20日

查看全文
  • Matilda Sipilä,
  • Farrokh Mehryary,
  • Sampo Pyysalo,
  • Filip Ginter,
  • Milica Todorović

摘要

Scientific text is a promising source of data in materials science, with ongoing research into utilising textual data for materials discovery. In this study, we developed and tested a Question Answering (QA) approach to extract material-property relationships from scientific publications. QA performance was evaluated for information extraction of perovskite bandgaps based on a human query. We observed considerable variation in results with five different large language models fine-tuned for the QA task. Best extraction accuracy was achieved with the QA MatSciBERT and F1-scores improved on the current state-of-the-art. QA also outperformed three latest generative large language models on the information extraction task, except the GPT-4 model. This work demonstrates the QA workflow and paves the way towards further applications. The simplicity and versatility of the QA approach all point to its considerable potential for text-driven discoveries in materials research.