3 papers
cond-mat.mtrl-sci2026
Understanding compression-extension error asymmetry in universal machine-learning interatomic potentials
Hongwei Du, Dingyang Lv, Boyan Zhang +11
Universal machine-learning interatomic potentials are increasingly used across diverse materials, yet their reliability under directional deformation remains unclear. We compare co…
cond-mat.mtrl-sci2026
Origins of Universal Machine Learning Force-Field Errors in Multicomponent Materials
Hongwei Du, Dingyang Lv, Baole Wei +11
Universal machine learning force-field generalization to multicomponent environments generated by compositional design remains insufficiently assessed. We construct a benchmark of…
cond-mat.mtrl-sci2026
A Hierarchical Synergistic Deep Learning Framework Integrating Composition, Structure, and Ionic Transport for Solid-State Electrolyte Discovery
Hongwei Du, Dingyang Lv, Baole Wei +5
Inorganic solid-state electrolytes must combine high room-temperature ionic conductivity, a wide electrochemical window, excellent electronic insulation, and favorable mechanical c…