Artificial Intelligence Framework for Scientific Information Extraction in Additive Manufacturing Literature

Strategic Insights -

Researchers at McGill University have made significant breakthroughs in developing a framework for collaborative information extraction from data-driven additive manufacturing (AM) literature using large language models (LLMs). This innovative approach aims to expedite the extraction of scientific information from AM and artificial intelligence (AI) contexts, reducing the need for manual effort. The proposed framework enables the collaboration between AM and AI experts to continuously extract relevant information from data-driven AM literature, demonstrating the potential for large language models to automate scientific information extraction.

Key Takeaways:

  • The research was funded by McGill Engineering Doctoral Award (MEDA) fellowship, National Research Council Canada, Graduate Excellence Award, and Mitacs Accelerate Program.
  • Over two dozen review articles contributed by AM domain experts have summarized scientific information from data-driven AM literature.
  • The proposed framework enables collaboration between AM and AI experts to extract scientific information from data-driven AM literature.
  • A demonstration tool was implemented based on the proposed framework, and a case study was conducted to extract information relevant to datasets, modeling, sensing, and AM system categories.
  • The research demonstrates the ability of large language models to expedite the extraction of relevant information from data-driven AM literature.
  • The framework has the potential to automate scientific information extraction, reducing manual effort required by experts.
  • Additional authors on the research include Mutahar Safdar, Jiarui Xie, and Andrei Mircea.

Statistics:

  • More than two dozen review articles contributed by AM domain experts have summarized scientific information from data-driven AM literature.
  • The proposed framework aims to automate the extraction of scientific information from data-driven AM literature, reducing manual effort by experts.
  • The research demonstrates the ability of large language models to expedite the extraction of relevant information from data-driven AM literature in 25(7) cases.

Sources:

  • NewsRx. Findings on Artificial Intelligence Reported by Investigators at McGill University (Human-artificial Intelligence Teaming for Scientific Information Extraction From Data-driven Additive Manufacturing Literature Using Large Language Models). Information Technology Newsweekly. July 15, 2025; p 256.
  • Human-artificial Intelligence Teaming for Scientific Information Extraction From Data-driven Additive Manufacturing Literature Using Large Language Models. Journal of Computing and Information Science in Engineering, 2025;25(7).