3 papers
cond-mat.mtrl-sci2026
Aligning Heterogeneous DFT Datasets: A Graph Neural Network Approach to Cross-Functional Formation Energies
Yidong Huang, Tenglong Lu, Hanwen Kang +3
Heterogeneous density functional theory (DFT) calculations, particularly plane-wave implementations, introduce systematic formation energy errors ranging from tens to hundreds of m…
cond-mat.mtrl-sci2026
Are Machine Learning Interatomic Potentials Truly Practical? A Benchmark of 23 Mainstream Models
Hanwen Kang, Tenglong Lu, Sheng Meng +1
Most MLIP benchmarks reward static accuracy while ignoring inference efficiency and hardware scalability -- driving model bloat with unclear real-world value. We benchmark 23 mains…
cond-mat.mtrl-sci2025
FastTrack: a fast method to evaluate mass transport in solid leveraging universal machine learning interatomic potential
Hanwen Kang, Tenglong Lu, Zhanbin Qi +3
We introduce a rapid, accurate framework for computing atomic migration barriers in crystals by combining universal machine learning force fields (MLFFs) with 3D potential energy s…