Reproduce “Automated Materials Data and Knowledge Extraction using Machine Learning and Large Language Models” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Aravindan Kamatchi Sundaram, B. Pabitramohan Prusty, Rohit Batra
The authors review automated methods for turning scattered materials-science literature into linked, machine-readable records. They trace approaches from rules and specialized NLP models to LLMs and agentic workflows that coordinate tools across text, tables and figures. On a synthesis-processing task, reported F1 scores range from 0.58 to 0.82–0.88. Yet complete sample-level records remain difficult, and errors limit database use.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.