Explainable Synthesizability Prediction of Inorganic Crystal Structures using Large Language Models

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 text-based models can be adapted to specialized cases where less data exists by transfer learning, demonstrated for the case of perovskites.