DERI Seminar with Dr Ricardo Grau-Crespo, Senior Lecturer of Computational Materials Science at QMUL
When: Thursday, November 27, 2025, 11:00 AM - 12:00 PM
Where: Teams
Speaker: Dr Ricardo Grau-Crespo, Senior Lecturer of Computational Materials Science at QM
Title: Crystal structure prediction with large language models
Abstract: The generation of plausible crystal structures is often the first step in predicting the structure and properties of a material from its chemical composition. However, physics-based methods for crystal structure prediction are computationally too expensive for high-throughput studies. I will introduce CrystaLLM, a methodology for the versatile generation of crystal structures, based on the autoregressive large language modeling (LLM) of the Crystallographic Information File (CIF) format. Trained on millions of CIF files, CrystaLLM focuses on modelling crystal structures through text. CrystaLLM can produce plausible crystal structures for a wide range of inorganic compounds unseen in training, as demonstrated by ab initio simulations. This work demonstrates the potential of LLMs for learning effective models of crystal chemistry. I will also discuss limitations of the model and ongoing work for its improvement and generalisation.
Bio: Dr Ricardo Grau-Crespo is a Senior Lecturer in Computational Materials Science at QMUL’s School of Engineering and Materials Science. He was born and raised in Cuba, and studied a BSc and MSc in Physics at the University of Havana. He did his PhD at Birkbeck, University of London (2006) followed by a postdoc at UCL. He got his first permanent academic position at the Chemistry Department of the University of Reading and moved to QMUL in June 2025. His research interests are in modelling materials, mainly for energy applications, using a combination of quantum chemistry and machine learning, and has published over 110 papers in this field. He is a Fellow of the Royal Society of Chemistry, and Associate Editor of the Journal of Materials Chemistry A and Materials advances.
________________________________________________________________________________
Microsoft Teams Need help?
Join the meeting now - Dr Ricardo Grau-Crespo, Senior Lecturer of Computational Materials Sciencespelli
Meeting ID: 394 435 684 304 84
Passcode: bj6ek2WZ