Yu-Tong Lu

LineShine: online acceleration for HPC-AI converged scientific computing

UniHam: A Large-Scale SOC-Complete Dataset and Benchmark for Hamiltonian Learning in Materials

Accurate prediction of electronic Hamiltonians would enable broad property inference while avoiding the high computational cost of Density Functional Theory (DFT). However, progress toward general-purpose materials foundation models is limited by a data bottleneck: existing Hamiltonian datasets are typically small, lac...

Evaluating Predicted Densities, Hamiltonians, and Density Matrices as Periodic SCF Initializers

Learned electronic states are usually evaluated by prediction error, even though their intended use is to accelerate the self-consistent-field (SCF) loop of density functional theory (DFT). We ask whether lower offline error actually yields a better SCF initializer. We construct $\rho$HD-43K, a 43,851-crystal DFT corpu...

Autism spectrum disorder burden across 798 locations, 1990–2023: frontier and inequality mapping in the context of child injury prevention and safety promotion

Introduction Autism spectrum disorder (ASD) is commonly identified in childhood and can affect communication, adaptive functioning, supervision, and access to health and educational services. These needs may be relevant to child safety planning, but this study did not measure injury outcomes. Methods We analyzed Global...

Anxiety and depressive disorders in adolescent injury prevention and safety promotion: frontier and inequality mapping across 953 locations

Purpose Anxiety and depressive disorders constitute major contributors to global mental health burdens, disproportionately affecting 10-24-year-olds during critical neurodevelopmental windows. Despite their substantial disease burden, persistent treatment gaps and inequitable resource allocation remain worldwide. This...

UniHam: A Large-Scale SOC-Complete Dataset and Benchmark for Hamiltonian Learning in Materials

Accurate prediction of electronic Hamiltonians would enable broad property inference while avoiding the high computational cost of Density Functional Theory (DFT). However, progress toward general-purpose materials foundation models is limited by a data bottleneck: existing Hamiltonian datasets are typically small, lac...