Universal machine-learning interatomic potentials (uMLIPs) are emerging as foundation models for atomistic simulation, offering near-ab initio accuracy at far lower cost. Their safe, broad deployment is limited by the absence of reliable, general uncertainty estimates. We present a unified, scalable uncertainty metric, U, built from a heterogeneous ensemble that reuses existing pretrained MLIPs. Across diverse chemistries and structures, U strongly tracks true prediction errors and robustly ranks configuration-level risk. Using U, we perform uncertainty-aware distillation to train system-specific potentials with far fewer labels: for tungsten, we match full density-functional-theory (DFT) training using 4% of the DFT data; for MoNbTaW, a dataset distilled by U supports high-accuracy potential training. By filtering numerical label noise, the distilled models can in some cases exceed the accuracy of the MLIPs trained on DFT data. This framework provides a practical reliability monitor and guides data selection and fine-tuning, enabling cost-efficient, accurate, and safer deployment of foundation models.

Iron oxidation is a complex process involving critical atomistic events, such as atomic adsorption, diffusion, and surface reconstruction, understanding of which is significant for both surface science and coating technology. Atomistic simulation serves as an useful tool to investigate the processes, where description of interatomic interactions is required. However, selecting appropriate force field or interatomic potential is not only difficult, but also essential for getting accurate result. In this work, we present a detailed benchmark of reactive force fields (ReaxFFs) and universal machine learning interatomic potentials (uMLIPs) against density functional theory (DFT) calculations of oxygen adsorption on various α-iron surfaces, which is the first yet crucial step towards oxidation. The comparisons show the coverage-dependent performance and improvable accuracy of both ReaxFFs and uMLIPs at reproducing DFT results, with ReaxFFs outperforming uMLIPs. Subsequently, iron oxidation is simulated using ReaxFF and uMLIP. The results reveal the strong capability of ReaxFF and poor stability of uMLIP for describing reactive process, i.e., the formation of iron oxide. This may be attributed to the suitable functional form of ReaxFF for the description of bond changes. The insights presented here not only provide an example of benchmarking force field or interatomic potential for system of interest, but also highlight the applicability of ReaxFF and scopes of improvement of uMLIP.

Recent advances in machine learning, combined with the generation of extensive density functional theory (DFT) datasets, have enabled the development of universal machine learning interatomic potentials (uMLIPs). These models offer broad applicability across the periodic table, achieving first-principles accuracy at a fraction of the computational cost of traditional DFT calculations. In this study, we demonstrate that state-of-the-art pretrained uMLIPs can effectively replace DFT for accurately modeling complex defects in a wide range of metals and alloys. Our investigation spans diverse scenarios, including grain boundaries and general defects in pure metals, defects in high-entropy alloys, hydrogen-alloy interactions, and solute-defect interactions. Remarkably, the latest EquiformerV2 models achieve DFT-level accuracy on comprehensive defect datasets, with root mean square errors below 5 meV atom−1 for energies and 100 meV Å−1 for forces, outperforming specialized machine learning potentials such as moment tensor potential and atomic cluster expansion. We also present a systematic analysis of accuracy versus computational cost and explore uncertainty quantification for uMLIPs. A detailed case study of tungsten (W) demonstrates that data on pure W alone is insufficient for modeling complex defects in uMLIPs, underscoring the critical importance of advanced machine learning architectures and diverse datasets, which include over 100 million structures spanning all elements. These findings establish uMLIPs as a robust alternative to DFT and a transformative tool for accelerating the discovery and design of high-performance materials.

Understanding atomic hydrogen (H) diffusion in multi-principal element alloys (MPEAs) is crucial for enhancing hydrogen transport and storage technologies. However, the vast compositional space and complex chemical environments of MPEAs pose significant challenges. We develop highly accurate machine learning force field and neural network-driven kinetic Monte Carlo simulations to investigate H diffusion in body-centered cubic (BCC) MoNbTaW MPEAs. H diffusion exhibits super-Arrhenius behavior in MPEAs, dominated by the low percentile of the H solution energy spectrum. Robust analytical models are derived via machine learning symbolic regression to predict H diffusivity across general BCC MPEAs. Additionally, it is revealed that chemical short-range order (SRO) generally does not impact H diffusion in MoNbTaW MPEAs, except it enhances diffusion when H-favoring elements are present in low concentrations. These insights not only deepen our understanding of H diffusion dynamics in MPEAs but also guide the strategic development of advanced MPEAs for hydrogen-related applications by manipulating element type, composition, and SRO.