Newest papers

Automated Materials Data and Knowledge Extraction using Machine Learning and Large Language Models

Arjun Sundaram, B. Pabitramohan Prusty, Rohit Batra · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.29895 · cited by 0 works (OpenAlex) · open-access PDF

Arabic

على مدى العقد الماضي، تم توجيه الجهود الكبيرة لتطوير أساليب التعلم الآلي (ML) لتخزين البيانات المادية وإمكانية اكتساب المعرفة المادية مباشرة من الكتابة العلمية. يتابع هذا التحليل تطور هذه الأساليب من قنوات البيانات القائمة على القواعد من خلال نموذج معالجة اللغة الطبيعية (NLP) ومركبات ذاتية التحويل إلى، في النهاية، نموذج اللغة الكبيرة (LLM) والأنظمة العاملية قادرة على استخراج المعلومات المادية عبر عدة طرق، بما في ذلك النصوص والمجلدات والأرقام. بالإضافة إلى بناء قواعد البيانات المادية الكبيرة والمتعددة التي يمكن الاستفادة منها مع المختبرات الذكية، ونحن نناقش كيف تمكنت هذه التقدمات من إنشاء القد

Machine translation (Meta M2M100); the English original follows.

Abstract

Over the past decade, substantial efforts have been devoted to developing machine learning (ML) methods for extracting materials data and, where possible, materials knowledge directly from the scientific literature. This review traces the evolution of these methods from heuristic, rule-based pipelines through natural language processing (NLP) models and transformer-based encoders to, lastly, large language models (LLMs) and agentic systems capable of extracting materials information across multiple modalities, including text, tables, and figures. Beside construction of large, queryable materials databases and knowledge graphs that support downstream materials discovery, we discuss how these advances have enabled new capabilities, such as, generic question answering across materials science, AI co-scientists and chatbots for research assistance, and closed-loop optimization and discovery when integrated with autonomous laboratories. We critically assess the datasets generated using these approaches, their extraction accuracy and computational costs, and the respective strengths and limitations of different methodologies, and even provide practical guidelines for developing more reliable and scalable materials data extraction pipelines. Finally, we identify key open challenges, including the extraction of full materials structure-property-processing-performance tetrahedron at the sample level, the scarcity of reliable benchmark datasets for systematic evaluation, and the resolution of non-standard materials terminology across different document modalities.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Computer science, Terminology, Artificial intelligence, Machine learning, Knowledge extraction, Information extraction

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange