{"paper_id":"3a5b0514-8a3f-4fce-bb89-b7d836cbf1ad","body_text":"A Graph Neural Network for the Era of Large Atomistic Models | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article A Graph Neural Network for the Era of Large Atomistic Models Han Wang, Duo Zhang, Anyang peng, Chun Cai, Wentao Li, Yuanchang Zhou, and 8 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7075747/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted You are reading this latest preprint version Abstract Foundation models, or large atomistic models (LAMs), aim to universally represent the ground-state potential energy surface (PES) of atomistic systems as defined by density functional theory (DFT). The scaling law is pivotal in the development of large models, suggesting that their generalizability in downstream tasks consistently improves with increased model size, expanded training datasets, and larger computational budgets. In this study, we present DPA3, a multi-layer graph neural network founded on line graph series (LiGS), designed explicitly for the era of LAMs. We demonstrate that the generalization error of the DPA3 model adheres to the scaling law. The scalability in the number of model parameters is attained by stacking additional layers within DPA3. Additionally, the model employs a dataset encoding mechanism that decouples the scaling of training data size from the model size within its multi-task training framework. When trained as problem-oriented potential energy models, the DPA3 model exhibits superior accuracy in the majority of benchmark cases, encompassing systems with diverse features, including molecules, bulk materials, surface and cluster catalysis, two-dimensional materials, and battery materials. When trained as a LAM on the OpenLAM-v1 dataset, the DPA-3.1-3M model exhibits state-of-the-art performance in the LAMBench benchmark suite for LAMs, demonstrating lowest overall zero-shot generalization error across 17 downstream tasks from a broad spectrum of research domains. This performance suggests superior accuracy as an out-of-the-box potential model, requiring minimal fine-tuning data for downstream scientific applications. Physical sciences/Materials science/Theory and computation/Atomistic models Physical sciences/Chemistry/Theoretical chemistry/Molecular dynamics Physical sciences/Physics/Chemical physics Full Text Additional Declarations There is NO Competing Interest. Cite Share Download PDF Status: Under Review Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {\"props\":{\"pageProps\":{\"initialData\":{\"identity\":\"rs-7075747\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":false,\"archivedVersions\":[],\"articleType\":\"Article\",\"associatedPublications\":[],\"authors\":[{\"id\":489751962,\"identity\":\"1ea19daf-4538-4f18-8081-80434f904c24\",\"order_by\":0,\"name\":\"Han Wang\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2klEQVRIiWNgGAWjYLACxgYGBn5mkrVINpOsxeAAsaoNjp89/IJxh12e8XHmZw9+5jDI8zcAGXi1nMlLs2A8k1xsdpjN3LB3G4PhjANs5gb4tJgdyDEzYGxjTtx2mMFMmnEbA+MGBh42Cbxazr8BaalP3NzM/g2kxZ6wlhs5xg8Y2w4nbmDmAduSSFCL/Y03ZgyMbccTZxzmKQf6RSJ5xmE2M7xaJPtzjD8wtlUn9vcf3/bg5zYb2/725md4tQABm/QfKAOIgYqJSAbMH2B6CasdBaNgFIyCEQkArD1DdBh3sQYAAAAASUVORK5CYII=\",\"orcid\":\"https://orcid.org/0000-0001-5623-1148\",\"institution\":\"Institute of Applied Physics and Computational Mathematics\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Han\",\"middleName\":\"\",\"lastName\":\"Wang\",\"suffix\":\"\"},{\"id\":489751963,\"identity\":\"61da03a2-f254-41a2-bff7-a33c10742bbf\",\"order_by\":1,\"name\":\"Duo Zhang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Peking University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Duo\",\"middleName\":\"\",\"lastName\":\"Zhang\",\"suffix\":\"\"},{\"id\":489751964,\"identity\":\"93f829d4-6bc7-48cf-be49-a347b0908069\",\"order_by\":2,\"name\":\"Anyang peng\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0002-0630-2187\",\"institution\":\"AI for Science Institute, Beijing\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Anyang\",\"middleName\":\"\",\"lastName\":\"peng\",\"suffix\":\"\"},{\"id\":489751965,\"identity\":\"bc7c9271-cb99-4d31-8701-5da3a35e162d\",\"order_by\":3,\"name\":\"Chun Cai\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"AI for Science Institute, Beijing\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Chun\",\"middleName\":\"\",\"lastName\":\"Cai\",\"suffix\":\"\"},{\"id\":489751966,\"identity\":\"27dbfd71-060f-44b8-bb19-8ab44dda0d4c\",\"order_by\":4,\"name\":\"Wentao Li\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Tsinghua University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Wentao\",\"middleName\":\"\",\"lastName\":\"Li\",\"suffix\":\"\"},{\"id\":489751967,\"identity\":\"b56551e4-f4a7-4241-90db-7be68ffd9a4f\",\"order_by\":5,\"name\":\"Yuanchang Zhou\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Chinese Academy of Sciences\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Yuanchang\",\"middleName\":\"\",\"lastName\":\"Zhou\",\"suffix\":\"\"},{\"id\":489751968,\"identity\":\"30e0b868-2ff8-45e9-b97a-150d7d280167\",\"order_by\":6,\"name\":\"Jinzhe Zeng\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0002-1515-8172\",\"institution\":\"University of Science and Technology of China\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Jinzhe\",\"middleName\":\"\",\"lastName\":\"Zeng\",\"suffix\":\"\"},{\"id\":489751969,\"identity\":\"65242508-579f-485d-93ce-fb2827fbbdb5\",\"order_by\":7,\"name\":\"Mingyu Guo\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"DP Technology\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Mingyu\",\"middleName\":\"\",\"lastName\":\"Guo\",\"suffix\":\"\"},{\"id\":489751970,\"identity\":\"07904844-e31c-4af3-a176-2fdde32fb9fc\",\"order_by\":8,\"name\":\"Chengqian Zhang\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Peking University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Chengqian\",\"middleName\":\"\",\"lastName\":\"Zhang\",\"suffix\":\"\"},{\"id\":489751971,\"identity\":\"e1a87aac-9b68-438c-8418-6f5c5068beed\",\"order_by\":9,\"name\":\"Bowen Li\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"East China Normal University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Bowen\",\"middleName\":\"\",\"lastName\":\"Li\",\"suffix\":\"\"},{\"id\":489751972,\"identity\":\"739eb4c6-ad9c-4404-9695-e67c41c2084b\",\"order_by\":10,\"name\":\"Hong Jiang\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0003-3187-2023\",\"institution\":\"College of Chemistry and Molecular Engineering, Beijing National Laboratory for Molecular Sciences, Peking University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Hong\",\"middleName\":\"\",\"lastName\":\"Jiang\",\"suffix\":\"\"},{\"id\":489751973,\"identity\":\"b6faa42f-37c6-45b3-8223-ee87dcc9a820\",\"order_by\":11,\"name\":\"Tong Zhu\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0001-7472-3736\",\"institution\":\"East China Normal University\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Tong\",\"middleName\":\"\",\"lastName\":\"Zhu\",\"suffix\":\"\"},{\"id\":489751974,\"identity\":\"f96e86bd-cd92-48cc-935a-8ce10dbe1bb5\",\"order_by\":12,\"name\":\"Weile Jia\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Chinese Academy of Sciences\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Weile\",\"middleName\":\"\",\"lastName\":\"Jia\",\"suffix\":\"\"},{\"id\":489751975,\"identity\":\"d01166f7-91b7-4fa8-b40c-bacd632ab6e1\",\"order_by\":13,\"name\":\"Linfeng Zhang\",\"email\":\"\",\"orcid\":\"https://orcid.org/0000-0002-8470-5846\",\"institution\":\"Princeton\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Linfeng\",\"middleName\":\"\",\"lastName\":\"Zhang\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2025-07-08 14:20:25\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-7075747/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-7075747/v1\",\"draftVersion\":[],\"editorialEvents\":[],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":89632401,\"identity\":\"ea36b725-6bfe-41da-b0a3-c0108ea1b493\",\"added_by\":\"auto\",\"created_at\":\"2025-08-22 06:53:53\",\"extension\":\"pdf\",\"order_by\":1,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":590076,\"visible\":true,\"origin\":\"\",\"legend\":\"Article File\",\"description\":\"\",\"filename\":\"DPA3manuscript.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-7075747/v1_covered_e4740e90-a8c7-4bce-8e26-0d0e3029603e.pdf\"}],\"financialInterests\":\"There is \\u003cb\\u003eNO\\u003c/b\\u003e Competing Interest.\",\"formattedTitle\":\"A Graph Neural Network for the Era of Large Atomistic Models\",\"fulltext\":[],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":false,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":true,\"hideJournal\":false,\"highlight\":\"\",\"institution\":\"\",\"isAcceptedByJournal\":false,\"isAuthorSuppliedPdf\":true,\"isDeskRejected\":\"\",\"isHiddenFromSearch\":false,\"isInQc\":false,\"isInWorkflow\":false,\"isPdf\":true,\"isPdfUpToDate\":true,\"isWithdrawnOrRetracted\":false,\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"nature-portfolio\",\"isNatureJournal\":true,\"hasQc\":false,\"allowDirectSubmit\":false,\"externalIdentity\":\"\",\"sideBox\":\"\",\"snPcode\":\"\",\"submissionUrl\":\"\",\"title\":\"Nature Portfolio\",\"twitterHandle\":\"\",\"acdcEnabled\":false,\"dfaEnabled\":false,\"editorialSystem\":\"ejp\",\"reportingPortfolio\":\"\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":false},\"keywords\":\"\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-7075747/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-7075747/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"Foundation models, or large atomistic models (LAMs), aim to universally represent the ground-state potential energy surface (PES) of atomistic systems as defined by density functional theory (DFT). The scaling law is pivotal in the development of large models, suggesting that their generalizability in downstream tasks consistently improves with increased model size, expanded training datasets, and larger computational budgets. In this study, we present DPA3, a multi-layer graph neural network founded on line graph series (LiGS), designed explicitly for the era of LAMs. We demonstrate that the generalization error of the DPA3 model adheres to the scaling law. The scalability in the number of model parameters is attained by stacking additional layers within DPA3. Additionally, the model employs a dataset encoding mechanism that decouples the scaling of training data size from the model size within its multi-task training framework. When trained as problem-oriented potential energy models, the DPA3 model exhibits superior accuracy in the majority of benchmark cases, encompassing systems with diverse features, including molecules, bulk materials, surface and cluster catalysis, two-dimensional materials, and battery materials. When trained as a LAM on the OpenLAM-v1 dataset, the DPA-3.1-3M model exhibits state-of-the-art performance in the LAMBench benchmark suite for LAMs, demonstrating lowest overall zero-shot generalization error across 17 downstream tasks from a broad spectrum of research domains. This performance suggests superior accuracy as an out-of-the-box potential model, requiring minimal fine-tuning data for downstream scientific applications.\",\"manuscriptTitle\":\"A Graph Neural Network for the Era of Large Atomistic Models\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2025-08-22 06:45:45\",\"doi\":\"10.21203/rs.3.rs-7075747/v1\",\"editorialEvents\":[],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"nature-communications\",\"isNatureJournal\":true,\"hasQc\":false,\"allowDirectSubmit\":false,\"externalIdentity\":\"NCOMMS\",\"sideBox\":\"Learn more about [Nature Communications](http://www.nature.com/ncomms/)\",\"snPcode\":\"\",\"submissionUrl\":\"https://mts-ncomms.nature.com/\",\"title\":\"Nature Communications\",\"twitterHandle\":\"\",\"acdcEnabled\":true,\"dfaEnabled\":true,\"editorialSystem\":\"ejp\",\"reportingPortfolio\":\"Nature Communications\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":false}}],\"origin\":\"\",\"ownerIdentity\":\"171bf724-9dfe-49c5-ab0f-2fe553379812\",\"owner\":[],\"postedDate\":\"August 22nd, 2025\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"under-review\",\"subjectAreas\":[{\"id\":51996454,\"name\":\"Physical sciences/Materials science/Theory and computation/Atomistic models\"},{\"id\":51996455,\"name\":\"Physical sciences/Chemistry/Theoretical chemistry/Molecular dynamics\"},{\"id\":51996456,\"name\":\"Physical sciences/Physics/Chemical physics\"}],\"tags\":[],\"updatedAt\":\"2025-08-22T06:45:45+00:00\",\"versionOfRecord\":[],\"versionCreatedAt\":\"2025-08-22 06:45:45\",\"video\":\"\",\"vorDoi\":\"\",\"vorDoiUrl\":\"\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-7075747\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-7075747\",\"identity\":\"rs-7075747\",\"version\":[\"v1\"]},\"buildId\":\"8U1c8b4HqxoKbykW_rLl7\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}