Nuclear masses at the predictive frontier

Nuclear masses for the next 25 years

Invited presentation on 05/2026

Mass is a fundamental observable in nuclear physics, encoding the integrated effect of nuclear interactions and anchoring a range of applications from astrophysics to tests of the Standard Model. Yet of the roughly 7,000 nuclei expected to be bound, only about half have ever been measured, leaving the rest to be predicted by theory. In this talk I survey today's modeling approaches, noting the rise of scientific machine learning as a complementary tool. I focus on how the field is moving from precise fits toward genuinely predictive, uncertainty-aware mass models. I close with a look ahead at how rapid integration of new measurements from facilities like FRIB will define the next decade of research with nuclear masses.