Benchmarking Universal Machine-Learning Interatomic Potentials for Temperature-Dependent Elasticity of Binary and High-Entropy Refractory Carbides
Abstract
Accurate predictions of temperature-dependent stiffness are essential for assessing the high-temperature performance of refractory carbides, yet the cost of ab initio molecular dynamics (AIMD) limits systematic exploration across compositions and temperatures. Universal machine-learning interatomic potentials (uMLIPs) offer a computationally efficient alternative, but their reliability for finite-temperature elasticity remains insufficiently established. Here, we benchmark nine uMLIPs against consistent AIMD reference data for five binary and two high-entropy (HE) carbides between 300 and 1200 K. Elastic constants and the corresponding bulk, shear, and Young's moduli are obtained using stress-strain molecular dynamics, explicitly sampling thermal atomic motion and anharmonic effects beyond thermal expansion alone. We assess absolute elastic properties and normalized thermal softening separately. MACE-MH-1 achieves the lowest overall mean absolute percentage error (5.7%). MACE-MH-1 and DPA4-Mini reproduce thermal softening most accurately, with mean deviations of 2.5 and 2.3 percentage points, respectively. All models generally underestimate stiffness, with larger equilibrium volumes relative to the AIMD reference likely contributing to this trend for most models. C12 exhibits the largest model-dependent errors. The HE carbides are described with accuracy comparable to that of the binary carbides, indicating no apparent accuracy penalty from chemical complexity. Accuracy instead varies with transition-metal composition, with group-V carbides, particularly TaC, presenting the greatest challenge. These results identify promising pretrained models for finite-temperature elasticity in carbides and show why accurate absolute stiffness and thermal softening must be assessed independently.
Related
- Benchmarking machine-learning interatomic potentials using classical force fields
- Benchmarking Machine-Learning Interatomic Potentials for Dynamical Stability in Inorganic Semiconductor Nanocrystals: A CdSe Case Study
- SLUSCHI-UP: A Web Infrastructure for SLUSCHI Melting-Temperature Calculations Using Universal Machine-Learning Interatomic Potentials
- Why Multi-Layer Message Passing Works: Completeness Theory for Graph Neural Network Interatomic Potentials
- MLIP Detective: Active Failure Mode Discovery Beyond Benchmark Scores for Machine-Learning Interatomic Potentials
BibTeX
@misc{lebeda2026benchmarking,
title = {{Benchmarking Universal Machine-Learning Interatomic Potentials for Temperature-Dependent Elasticity of Binary and High-Entropy Refractory Carbides}},
author = {Miroslav Lebeda and Jan Drahokoupil and Šimon Svoboda and Petr Vlčák},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.11878},
eprint = {2610.11878},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.11878},
}
RIS
TY - UNPB
TI - Benchmarking Universal Machine-Learning Interatomic Potentials for Temperature-Dependent Elasticity of Binary and High-Entropy Refractory Carbides
AU - Miroslav Lebeda
AU - Jan Drahokoupil
AU - Šimon Svoboda
AU - Petr Vlčák
PY - 2026
DA - 2026/10/28
PB - arXiv
DO - 10.48550/arxiv.2610.11878
UR - https://arxiv.org/abs/2610.11878
AB - Accurate predictions of temperature-dependent stiffness are essential for assessing the high-temperature performance of refractory carbides, yet the cost of ab initio molecular dynamics (AIMD) limits systematic exploration across compositions and temperatures. Universal machine-learning interatomic potentials (uMLIPs) offer a computationally efficient alternative, but their reliability for finite-temperature elasticity remains insufficiently established. Here, we benchmark nine uMLIPs against consistent AIMD reference data for five binary and two high-entropy (HE) carbides between 300 and 1200 K. Elastic constants and the corresponding bulk, shear, and Young's moduli are obtained using stress-strain molecular dynamics, explicitly sampling thermal atomic motion and anharmonic effects beyond thermal expansion alone. We assess absolute elastic properties and normalized thermal softening separately. MACE-MH-1 achieves the lowest overall mean absolute percentage error (5.7%). MACE-MH-1 and DPA4-Mini reproduce thermal softening most accurately, with mean deviations of 2.5 and 2.3 percentage points, respectively. All models generally underestimate stiffness, with larger equilibrium volumes relative to the AIMD reference likely contributing to this trend for most models. C12 exhibits the largest model-dependent errors. The HE carbides are described with accuracy comparable to that of the binary carbides, indicating no apparent accuracy penalty from chemical complexity. Accuracy instead varies with transition-metal composition, with group-V carbides, particularly TaC, presenting the greatest challenge. These results identify promising pretrained models for finite-temperature elasticity in carbides and show why accurate absolute stiffness and thermal softening must be assessed independently.
ER -