{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T19:44:29Z","timestamp":1780429469690,"version":"3.54.1"},"reference-count":53,"publisher":"IOP Publishing","issue":"3","license":[{"start":{"date-parts":[[2025,7,21]],"date-time":"2025-07-21T00:00:00Z","timestamp":1753056000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2025,7,21]],"date-time":"2025-07-21T00:00:00Z","timestamp":1753056000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Mach. Learn.: Sci. Technol."],"published-print":{"date-parts":[[2025,9,30]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>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<jats:sup>\u22121<\/jats:sup> for energies and 100 meV \u00c5<jats:sup>\u22121<\/jats:sup> 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.<\/jats:p>","DOI":"10.1088\/2632-2153\/adea2d","type":"journal-article","created":{"date-parts":[[2025,6,30]],"date-time":"2025-06-30T18:58:29Z","timestamp":1751309909000},"page":"030501","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":8,"title":["Universal machine learning interatomic potentials poised to supplant DFT in modeling general defects in metals and random alloys"],"prefix":"10.1088","volume":"6","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6333-9237","authenticated-orcid":true,"given":"Fei","family":"Shuang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-3900-3995","authenticated-orcid":true,"given":"Zixiong","family":"Wei","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kai","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wei","family":"Gao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Poulumi","family":"Dey","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"266","published-online":{"date-parts":[[2025,7,21]]},"reference":[{"key":"mlstadea2dbib1","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-024-54554-x","article-title":"General-purpose machine-learned potential for 16 elemental metals and their alloys","volume":"15","author":"Song","year":"2024","journal-title":"Nat. Commun."},{"key":"mlstadea2dbib2","doi-asserted-by":"publisher","first-page":"624","DOI":"10.1038\/s41586-025-08628-5","article-title":"A generative model for inorganic materials design","volume":"639","author":"Zeni","year":"2025","journal-title":"Nature"},{"key":"mlstadea2dbib3","article-title":"Matbench discovery\u2014a framework to evaluate machine learning crystal stability predictions","author":"Riebesell","year":"2023"},{"key":"mlstadea2dbib4","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevB.96.024104","article-title":"Including crystal structure attributes in machine learning models of formation energies via Voronoi tessellations","volume":"96","author":"Ward","year":"2017","journal-title":"Phys. Rev. B"},{"key":"mlstadea2dbib5","doi-asserted-by":"publisher","first-page":"126","DOI":"10.1016\/j.mattod.2021.08.012","article-title":"Accelerating materials discovery with Bayesian optimization and graph deep learning","volume":"51","author":"Zuo","year":"2021","journal-title":"Mater. Today"},{"key":"mlstadea2dbib6","doi-asserted-by":"publisher","first-page":"4117","DOI":"10.1126\/sciadv.abn4117","article-title":"Rapid discovery of stable materials by coordinate-free coarse graining","volume":"8","author":"Goodall","year":"2022","journal-title":"Sci. Adv."},{"key":"mlstadea2dbib7","doi-asserted-by":"publisher","first-page":"211","DOI":"10.1038\/s41524-022-00891-8","article-title":"Data-augmentation for graph neural network learning of the relaxed energies of unrelaxed structures","volume":"8","author":"Gibson","year":"2022","journal-title":"npj Comput. Mater."},{"key":"mlstadea2dbib8","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.120.145301","article-title":"Crystal graph convolutional neural networks for an accurate and interpretable prediction of material properties","volume":"120","author":"Xie","year":"2018","journal-title":"Phys. Rev. Lett."},{"key":"mlstadea2dbib9","doi-asserted-by":"publisher","first-page":"3564","DOI":"10.1021\/acs.chemmater.9b01294","article-title":"Graph networks as a universal machine learning framework for molecules and crystals","volume":"31","author":"Chen","year":"2019","journal-title":"Chem. Mater."},{"key":"mlstadea2dbib10","doi-asserted-by":"publisher","first-page":"185","DOI":"10.1038\/s41524-021-00650-1","article-title":"Atomistic line graph neural network for improved materials property predictions","volume":"7","author":"Choudhary","year":"2021","journal-title":"npj Comput. Mater."},{"key":"mlstadea2dbib11","doi-asserted-by":"publisher","first-page":"718","DOI":"10.1038\/s43588-022-00349-3","article-title":"A universal graph deep learning interatomic potential for the periodic table","volume":"2","author":"Chen","year":"2022","journal-title":"Nat. Comput. Sci."},{"key":"mlstadea2dbib12","doi-asserted-by":"publisher","first-page":"1031","DOI":"10.1038\/s42256-023-00716-3","article-title":"CHGNet as a pretrained universal neural network potential for charge-informed atomistic modelling","volume":"5","author":"Deng","year":"2023","journal-title":"Nat. Mach. Intell."},{"key":"mlstadea2dbib13","article-title":"A foundation model for atomistic materials chemistry","author":"Batatia","year":"2023"},{"key":"mlstadea2dbib14","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevX.14.021036","article-title":"Graph atomic cluster expansion for semilocal interactions beyond equivariant message passing","volume":"14","author":"Bochkarev","year":"2024","journal-title":"Phys. Rev. X"},{"key":"mlstadea2dbib15","doi-asserted-by":"publisher","first-page":"4857","DOI":"10.1021\/acs.jctc.4c00190","article-title":"Scalable parallel algorithm for graph neural network interatomic potentials in molecular dynamics simulations","volume":"20","author":"Park","year":"2024","journal-title":"J. Chem. Theory Comput."},{"key":"mlstadea2dbib16","article-title":"Orb: a fast, scalable neural network potential","author":"Neumann","year":"2024"},{"key":"mlstadea2dbib17","doi-asserted-by":"publisher","first-page":"80","DOI":"10.1038\/s41586-023-06735-9","article-title":"Scaling deep learning for materials discovery","volume":"624","author":"Merchant","year":"2023","journal-title":"Nature"},{"key":"mlstadea2dbib18","article-title":"MatterSim: a deep learning atomistic model across elements, temperatures and pressures","author":"Yang","year":"2024"},{"key":"mlstadea2dbib19","article-title":"Open materials 2024 (OMat24) inorganic materials dataset and models","author":"Barroso-Luque","year":"2024"},{"key":"mlstadea2dbib20","article-title":"EquiformerV2: improved equivariant transformer for scaling to higher-degree representations 12th Int. Conf. on Learning Representations, ICLR 2024","author":"Liao","year":"2023"},{"key":"mlstadea2dbib21","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1038\/s41524-024-01493-2","article-title":"DPA-2: a large atomic model as a multi-task learner","volume":"10","author":"Zhang","year":"2024","journal-title":"npj Comput. Mater."},{"key":"mlstadea2dbib22","doi-asserted-by":"publisher","DOI":"10.1063\/5.0155600","article-title":"DeePMD-kit v2: a software package for deep potential models","volume":"159","author":"Zeng","year":"2023","journal-title":"J. Chem. Phys."},{"key":"mlstadea2dbib23","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1038\/s41524-024-01500-6","article-title":"Systematic softening in universal machine learning interatomic potentials","volume":"11","author":"Deng","year":"2025","journal-title":"npj Comput. Mater."},{"key":"mlstadea2dbib24","doi-asserted-by":"publisher","first-page":"e58","DOI":"10.1002\/mgea.58","article-title":"Systematic assessment of various universal machine\u2010learning interatomic potentials","volume":"2","author":"Yu","year":"2024","journal-title":"Mater. Genome Eng. Adv."},{"key":"mlstadea2dbib25","doi-asserted-by":"publisher","first-page":"13111","DOI":"10.1021\/acsami.4c03815","article-title":"Performance assessment of universal machine learning interatomic potentials: challenges and directions for materials\u2019 surfaces","volume":"17","author":"Focassio","year":"2024","journal-title":"ACS Appl. Mater. Interfaces"},{"key":"mlstadea2dbib26","doi-asserted-by":"publisher","first-page":"15","DOI":"10.1016\/0927-0256(96)00008-0","article-title":"Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set","volume":"6","author":"Kresse","year":"1996","journal-title":"Comput. Mater. Sci."},{"key":"mlstadea2dbib27","doi-asserted-by":"publisher","first-page":"3865","DOI":"10.1103\/PhysRevLett.77.3865","article-title":"Generalized gradient approximation made simple","volume":"77","author":"Perdew","year":"1996","journal-title":"Phys. Rev. Lett."},{"key":"mlstadea2dbib28","doi-asserted-by":"publisher","first-page":"17953","DOI":"10.1103\/PhysRevB.50.17953","article-title":"Projector augmented-wave method","volume":"50","author":"Bl\u00f6chl","year":"1994","journal-title":"Phys. Rev. B"},{"key":"mlstadea2dbib29","doi-asserted-by":"publisher","DOI":"10.1016\/j.cpc.2021.108033","article-title":"VASPKIT: a user-friendly interface facilitating high-throughput computing and analysis using VASP code","volume":"267","author":"Wang","year":"2021","journal-title":"Comput. Phys. Commun."},{"key":"mlstadea2dbib30","doi-asserted-by":"publisher","DOI":"10.1088\/1361-648X\/aa680e","article-title":"The atomic simulation environment\u2014a Python library for working with atoms","volume":"29","author":"Hjorth Larsen","year":"2017","journal-title":"J. Phys.: Condens. Matter"},{"key":"mlstadea2dbib31","doi-asserted-by":"publisher","DOI":"10.1088\/0965-0393\/18\/1\/015012","article-title":"Visualization and analysis of atomistic simulation data with OVITO\u2013the open visualization tool","volume":"18","author":"Stukowski","year":"2010","journal-title":"Model. Simul. Mater. Sci. Eng."},{"key":"mlstadea2dbib32","doi-asserted-by":"publisher","DOI":"10.1016\/j.mtphys.2024.101560","article-title":"Improving machine-learning models in materials science through large datasets","volume":"48","author":"Schmidt","year":"2024","journal-title":"Mater. Today Phys."},{"key":"mlstadea2dbib33","article-title":"Generalizing denoising to non-equilibrium structures improves equivariant force fields","author":"Liao","year":"2024"},{"key":"mlstadea2dbib34","doi-asserted-by":"publisher","first-page":"40","DOI":"10.1016\/j.actamat.2019.12.030","article-title":"Grain boundary properties of elemental metals","volume":"186","author":"Zheng","year":"2020","journal-title":"Acta Mater."},{"key":"mlstadea2dbib35","doi-asserted-by":"publisher","first-page":"118","DOI":"10.1038\/s41524-025-01599-1","article-title":"Modeling extensive defects in metals through classical potential-guided sampling and automated configuration reconstruction","volume":"11","author":"Shuang","year":"2025","journal-title":"npj Comput. Mater."},{"key":"mlstadea2dbib36","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevB.107.104103","article-title":"Systematic atomic structure datasets for machine learning potentials: application to defects in magnesium","volume":"107","author":"Poul","year":"2023","journal-title":"Phys. Rev. B"},{"key":"mlstadea2dbib37","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.2322962121","article-title":"Quantifying chemical short-range order in metallic alloys","volume":"121","author":"Sheriff","year":"2024","journal-title":"Proc. Natl Acad. Sci."},{"key":"mlstadea2dbib38","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijplas.2025.104308","article-title":"Size-dependent strength superiority in multi-principal element alloys versus constituent metals: insights from machine-learning atomistic simulations","volume":"188","author":"Shuang","year":"2025","journal-title":"Int. J. Plast."},{"key":"mlstadea2dbib39","doi-asserted-by":"publisher","first-page":"4484","DOI":"10.1038\/s41467-019-12452-7","article-title":"Local electronic descriptors for solute-defect interactions in bcc refractory metals","volume":"10","author":"Hu","year":"2019","journal-title":"Nat. Commun."},{"key":"mlstadea2dbib40","doi-asserted-by":"publisher","first-page":"178","DOI":"10.1038\/s41524-025-01650-1","article-title":"Universal machine learning interatomic potentials are ready for phonons","volume":"11","author":"Loew","year":"2025","journal-title":"npj Comput. Mater."},{"key":"mlstadea2dbib41","doi-asserted-by":"publisher","first-page":"2105","DOI":"10.1021\/acsmaterialslett.5c00093","article-title":"CHIPS-FF: evaluating universal machine learning force fields for material properties","volume":"7","author":"Wines","year":"2025","journal-title":"ACS Mater. Lett."},{"key":"mlstadea2dbib42","doi-asserted-by":"publisher","first-page":"510","DOI":"10.1557\/s43579-022-00221-5","article-title":"Machine-learning potentials for crystal defects","volume":"12","author":"Freitas","year":"2022","journal-title":"MRS Commun."},{"key":"mlstadea2dbib43","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevB.69.144113","article-title":"Misfit-energy-increasing dislocations in vapor-deposited CoFe\/NiFe multilayers","volume":"69","author":"Zhou","year":"2004","journal-title":"Phys. Rev. B"},{"key":"mlstadea2dbib44","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevMaterials.7.043801","article-title":"Active learning strategies for atomic cluster expansion models","volume":"7","author":"Lysogorskiy","year":"2023","journal-title":"Phys. Rev. Mater."},{"key":"mlstadea2dbib45","doi-asserted-by":"publisher","DOI":"10.1016\/j.cpc.2019.106949","article-title":"DScribe: library of descriptors for machine learning in materials science","volume":"247","author":"Himanen","year":"2020","journal-title":"Comput. Phys. Commun."},{"key":"mlstadea2dbib46","doi-asserted-by":"publisher","DOI":"10.1063\/5.0234287","article-title":"How graph neural network interatomic potentials extrapolate: role of the message-passing algorithm","volume":"161","author":"Kang","year":"2024","journal-title":"J. Chem. Phys."},{"key":"mlstadea2dbib47","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevMaterials.7.045802","article-title":"Modeling high-entropy transition metal alloys with alchemical compression","volume":"7","author":"Lopanitsyna","year":"2023","journal-title":"Phys. Rev. Mater."},{"key":"mlstadea2dbib48","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.131.028001","article-title":"Tensor-reduced atomic density representations","volume":"131","author":"Darby","year":"2023","journal-title":"Phys. Rev. Lett."},{"key":"mlstadea2dbib49","article-title":"The importance of being scalable: improving the speed and accuracy of neural network interatomic potentials across chemical domains","author":"Qu","year":"2024"},{"key":"mlstadea2dbib50","article-title":"Specialized machine learning force fields: distilling foundation models via energy hessians","author":"Amin","year":"2025"},{"key":"mlstadea2dbib51","article-title":"Forces are not enough: benchmark and critical evaluation for machine learning force fields with molecular simulations","author":"Fu","year":"2022"},{"key":"mlstadea2dbib52","article-title":"The dark side of the forces: assessing non-conservative force models for atomistic machine learning","author":"Bigi","year":"2024"},{"key":"mlstadea2dbib53","article-title":"Data For: Universal machine learning interatomic potentials poised to supplant DFT in modeling general defects in metals and random alloys GitHub","author":"Shuang","year":"2025"}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adea2d","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adea2d\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adea2d","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adea2d\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adea2d\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adea2d\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adea2d\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adea2d\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,21]],"date-time":"2025-07-21T10:14:48Z","timestamp":1753092888000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/adea2d"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,21]]},"references-count":53,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2025,7,21]]},"published-print":{"date-parts":[[2025,9,30]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/adea2d","relation":{},"ISSN":["2632-2153"],"issn-type":[{"value":"2632-2153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7,21]]},"assertion":[{"value":"Universal machine learning interatomic potentials poised to supplant DFT in modeling general defects in metals and random alloys","name":"article_title","label":"Article Title"},{"value":"Machine Learning: Science and Technology","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2025 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2025-02-21","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-06-30","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-07-21","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}