DOE OSTI · 2536774
Explainable Synthesizability Prediction of Inorganic Crystal Polymorphs Using Large Language Models
Abstract
Abstract We evaluate the ability of machine learning to predict whether a hypothetical crystal structure can be synthesized and explain those predictions to scientists. Fine‐tuned large language models (LLMs) trained on a human‐readable text description of the target crystal structure perform comparably to previous bespoke convolutional graph neural network methods, but better prediction quality can be achieved by training a positive‐unlabeled learning model on a text‐embedding representation of the structure. An LLM‐based workflow can then be used to generate human‐readable explanations for the types of factors governing synthesizability, extract the underlying physical rules, and assess the veracity of those rules. These explanations can guide chemists in modifying or optimizing non‐synthesizable hypothetical structures to make them more feasible for materials design.
Keep this discovery
Explore connections, maps & timelines
Kim, Seongmin [Department of Chemical and Biological Engineering (BK21 four) Seoul National University 1 Gwanak‐ro Gwanak‐gu Seoul 08826 South Korea, Department of Chemical and Biomolecular Engineering Korea Advanced Institute of Science and Technology (KAIST) 291, Daehak‐ro Yuseong‐gu Daejeon 34141 South Korea], Schrier, Joshua [Department of Chemistry and Biochemistry Fordham University 441 E. Fordham Road The Bronx New York 10458 USA], Jung, Yousung [Department of Chemical and Biological Engineering (BK21 four) Seoul National University 1 Gwanak‐ro Gwanak‐gu Seoul 08826 South Korea, Institute of Chemical Processes Seoul National University 1 Gwanak‐ro Gwanak‐gu Seoul 08826 South Korea, Institute of Engineering Research Seoul National University 1 Gwanak‐ro Gwanak‐gu Seoul 08826 South Korea]. 2025-03-22. Explainable Synthesizability Prediction of Inorganic Crystal Polymorphs Using Large Language Models. https://doi.org/10.1002/ange.202423950
Cite the original work for its findings. Save a collection to share your selection of sources.