Abstract
Molecular property prediction is a fundamental task in the drug and materialindustries. Physically, the properties of a molecule are determined by its ownelectronic structure, which can be exactly described by the Schr\"odingerequation. However, solving the Schr\"odinger equation for most molecules isextremely challenging due to long-range interactions in the behavior of aquantum many-body system. While deep learning methods have proven to beeffective in molecular property prediction, we design a novel method, namelyGEM-2, which comprehensively considers both the long-range and many-bodyinteractions in molecules. GEM-2 consists of two interacted tracks: anatom-level track modeling both the local and global correlation between any twoatoms, and a pair-level track modeling the correlation between all atom pairs,which embed information between any 3 or 4 atoms. Extensive experimentsdemonstrated the superiority of GEM-2 over multiple baseline methods in quantumchemistry and drug discovery tasks.