AI RESEARCH
PFT: Phonon Fine-tuning for Machine Learned Interatomic Potentials
arXiv CS.LG
•
ArXi:2601.07742v4 Announce Type: replace-cross Many materials properties depend on higher-order derivatives of the potential energy surface, yet machine learned interatomic potentials (MLIPs) trained with a standard loss on energy, force, and stress errors can exhibit error in curvature, degrading the prediction of vibrational properties. We