A machine learning tight-binding framework reconstructs DFT-level Hamiltonians and computes electronic properties for systems with up to 100 million atoms, including graphene mobility versus carrier concentration.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cond-mat.mtrl-sci 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
GPUTB: Efficient Machine Learning Tight-Binding Method for Large-Scale Electronic Properties Calculations
A machine learning tight-binding framework reconstructs DFT-level Hamiltonians and computes electronic properties for systems with up to 100 million atoms, including graphene mobility versus carrier concentration.