Large language models (LLMs) are increasingly being used in materials science. However, little attention has been given to benchmarking and standardized evaluation for LLM-based materials property prediction, which hinders progress. We present LLM4Mat-Bench, the largest benchmark to date for evaluating the performance of LLMs in predicting the properties of crystalline materials. LLM4Mat-Bench contains about 1.9M crystal structures in total, collected from 10 publicly available materials data sources, and 45 distinct properties. LLM4Mat-Bench features different input modalities: crystal composition, CIF, and crystal text description, with 4.7M, 615.5M, and 3.1B tokens in total for each modality, respectively. We use LLM4Mat-Bench to fine-tune models with different sizes, including LLM-Prop and MatBERT, and provide zero-shot and few-shot prompts to evaluate the property prediction capabilities of LLM-chat-like models, including Llama, Gemma, and Mistral. The results highlight the challenges of general-purpose LLMs in materials science and the need for task-specific predictive models and task-specific instruction-tuned LLMs in materials property prediction.
@article{arxiv.2411.00177,
title = {LLM4Mat-Bench: Benchmarking Large Language Models for Materials Property Prediction},
author = {Andre Niyongabo Rubungo and Kangming Li and Jason Hattrick-Simpers and Adji Bousso Dieng},
journal= {arXiv preprint arXiv:2411.00177},
year = {2024}
}
Comments
Accepted at NeurIPS 2024-AI4Mat Workshop. The Benchmark and code can be found at https://github.com/vertaix/LLM4Mat-Bench