Author(s): Zilong Yuan, Zechen Tang, Honggeng Tao, Xiaoxun Gong, Zezhou Chen, Yuxiang Wang, He Li, Yang Li, Zhiming Xu, Minghui Sun, Boheng Zhao, Chen Si, Chong Wang, Wenhui Duan, and Yong Xu Integrating essential physical priors into deep-learning first-principles methods is a critical fundamental problem. Here we demonstrate that the deep learning density functional theory Hamiltonian (DeepH) method can be substantially improved by changing the learning objective to a rotation-invarian… [Phys. Rev. Lett. 137, 046401] Published Mon Jul 20, 2026