2 papers
cond-mat.mtrl-sci2026
Machine Learning Materials Properties by Encoding Orbital-Projected Density of States
Paulo Pires, Pierre-Paul De Breuck, Mauro Fava +2
Graph neural networks have become the dominant machine-learning architecture for predicting materials properties from crystal structures. Yet the initialization of atomic node feat…
cond-mat.mtrl-sci2025
Universal Machine Learning Interatomic Potentials are Ready for Phonons
Antoine Loew, Dewen Sun, Hai-Chen Wang +2
There has been an ongoing race for the past several years to develop the best universal machinelearning interatomic potential. This progress has led to increasingly accurate models…