[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120251-en":3,"doc-seo-120251-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},120251,7971461740909,"Levi","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","Systematic softening in universal machine learning interatomic potentials - read online","Machine learning interatomic potentials (MLIPs) enable large-scale atomic simulations, but reliable out-of-distribution extrapolation remains uncertain for universal MLIPs (uMLIPs). This study identifies a consistent potential energy surface (PES) softening effect across M3GNet, CHGNet, and MACE-MP-0, showing energy and force underprediction in benchmark tasks spanning surfaces, defects, solid-solution energetics, ion migration barriers, phonon modes, and high-energy states. The effect is traced to systematically underestimated PES curvature caused by biased near-equilibrium sampling in uMLIP pre-training datasets.","UC Berkeley  \nUC Berkeley Previously Published Works  \nTitle  \nSystematic softening in universal machine learning interatomic potentials  \nPermalink  \n[https://escholarship.org/uc/item/7zz1s3tj](https://escholarship.org/uc/item/7zz1s3tj)  \nJournal  \nnpj Computational Materials, 11(1)  \nISSN  \n2057-3960  \nAuthors  \nDeng, Bowen  \nChoi, Yunyeong Zhong, Peichenet al.  \nPublication Date  \n2025  \nDOI  \n10.1038/s41524-024-01500-6  \nCopyright Information  \nThis work is made available under the terms of a Creative Commons Attribution License, available at [https://creativecommons.org/licenses/by/4.0/](https://creativecommons.org/licenses/by/4.0/)  \nPeer reviewed  \n[eScholarship.org](eScholarship.org) Powered by the California Digital Library  \nUniversity of California  \nnpj | computational materials Article  \nPublished in partnership with the Shanghai Institute of Ceramics of the Chinese Academy of Sciences  \n[https://doi.org/10.1038/s41524-024-01500-6](https://doi.org/10.1038/s41524-024-01500-6)  \nSystematic softening in universal machine learning interatomic potentials  \n Check for updates  \n\n| Bowen Deng 1,2, Yunyeong Choi 1,2, Peichen Zhong 1,2, Janosh Riebesell 3, Shashwat Anand  Zhuohan Li 2, KyuJung Jun 1,2, Kristin A. Persson 1,2 & Gerbrand Ceder 1,2  |  |  | 2\u003Cbr>, |\n| --- | --- | --- | --- |\n| Machine learning interatomic potentials (MLIPs) have introduced a new paradigm for atomic simulations. Recent advancements have led to universal MLIPs (uMLIPs) that are pre-trained on diverse datasets, providing opportunities for universal force fields and foundational machine learning models. However, their performance in extrapolating to out-of-distribution complex atomic environments remains unclear. In this study, we highlight a consistent potential energy surface (PES) softening effect in three uMLIPs: M3GNet, CHGNet, and MACE-MP-0, which is characterized by energy and force underprediction in atomic-modeling benchmarks including surfaces, defects, solidsolution energetics, ion migration barriers, phonon vibration modes, and general high-energy states. The PES softening behavior originates primarily from the systematically underpredicted PES curvature, which derives from the biased sampling of near-equilibrium atomic arrangements in uMLIP pre-training datasets. Our findings suggest that a considerable fraction of uMLIP errors are highly systematic, and can therefore be efficiently corrected. We argue for the importance of a comprehensive materials dataset with improved PES sampling for next-generation\u003Cbr>foundational MLIPs. |  |  |  |\n| Artificial intelligence (AI) is increasingly shifting the paradigm ofscientific discovery to accelerate research and solve real-world scientific challenges1. While ab-initio quantum mechanical simulation methods, such as density functional theory (DFT), offer the theoretical foundation to investigate material and chemical science problems at the atomic scale, their computational demands limit their applicability in both spatial and temporal scales. Recent advancementsin machine learning interatomic potentials(MLIPs)2,3 have enabled the opportunity to scale up quantum mechanical methods to million atoms simulations such as water, copper4, and biomolecules5.\u003Cbr>Alongside improvements in atomic environment descriptors and graph neural networks that enhance the expressivity of MLIP models3,6, universal machine learning interatomic potentials (uMLIPs) have demonstrated another avenue by taking advantage of pre-training on large and comprehensive material datasets7–13. These uMLIPs enable out-of-box atomic modeling covering the entire periodic table as well as providing robust machine-learning foundations for fine-tuning downstream tasks. While uMLIPs hold considerable promise, a critical challenge lies in their ability to reliably generalize to complex and diverse chemical environments, particularly those that deviate significantly from the pre-training data distribution. Several recent benchmark efforts have te","cbCairUD9M0HNLtB","https://ap.wps.com/l/cbCairUD9M0HNLtB","pdf",3872986,1,10,"English","en",105,"# Abstract\n# Background and motivation\n# Study scope and uMLIP models\n# Benchmark results\n## Energy and force underprediction\n# Mechanism analysis\n## Biased near-equilibrium sampling\n# Implications for next-generation foundational MLIPs","[{\"question\":\"What is the main finding about universal MLIPs in this study?\",\"answer\":\"The study reports a consistent potential energy surface (PES) softening effect across multiple uMLIPs, leading to systematic underprediction of energies and forces on benchmarks.\"},{\"question\":\"Which uMLIP models are analyzed in the paper?\",\"answer\":\"The work studies three universal MLIPs: M3GNet, CHGNet, and MACE-MP-0 (referred to as MACE in the text).\"},{\"question\":\"Why does PES softening occur according to the paper?\",\"answer\":\"The authors attribute it to systematically underestimated PES curvature stemming from biased sampling of near-equilibrium atomic configurations in the pre-training datasets.\"}]","Systematic softening in universal machine learning interatomic potentials - read online | PDF",1785729018,25,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"systematic-softening-in-universal-machine-learning-interatomic-potentials-read-online","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/systematic-softening-in-universal-machine-learning-interatomic-potentials-read-online/120251/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What is the main finding about universal MLIPs in this study?","Question",{"text":75,"@type":76},"The study reports a consistent potential energy surface (PES) softening effect across multiple uMLIPs, leading to systematic underprediction of energies and forces on benchmarks.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Which uMLIP models are analyzed in the paper?",{"text":80,"@type":76},"The work studies three universal MLIPs: M3GNet, CHGNet, and MACE-MP-0 (referred to as MACE in the text).",{"name":82,"@type":73,"acceptedAnswer":83},"Why does PES softening occur according to the paper?",{"text":84,"@type":76},"The authors attribute it to systematically underestimated PES curvature stemming from biased sampling of near-equilibrium atomic configurations in the pre-training datasets.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,134],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":21,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":21,"slug":133},"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]