Benchmarking bandgap prediction in semiconductors under experimental and realistic evaluation settings

Benchmarking bandgap prediction in semiconductors under experimental and realistic evaluation settings | 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 Resource Benchmarking bandgap prediction in semiconductors under experimental and realistic evaluation settings Haiping Lu, Haolin Wang, Xianyuan Liu, Anna Jungbluth, Alexandra Ramadan, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9453126/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 Accurate bandgap prediction is crucial for semiconductor applications, yet machine learning models trained on computational data often struggle to generalize to experimental bandgap measurements. Challenges related to data fidelity, domain generalization, and model interpretability remain insufficiently addressed in existing evaluation frameworks. To bridge this gap, we introduce RealMat-BaG, a benchmark for assessing model reliability under experimentally relevant conditions. We curate an open-access dataset of experimental bandgaps with aligned crystal structures and compare graph neural networks as well as classical machine learning baselines. Our framework evaluates performance across statistical and domain-based splits, examines transfer from DFT-computed to experimental bandgaps, and analyzes interpretability at both elemental-property and structural levels. Our results reveal the fundamental generalization limitations of current bandgap prediction models and establish a benchmark aligned with experimental measurements for developing more reliable learning strategies for materials discovery. Physical sciences/Mathematics and computing/Computer science Physical sciences/Materials science/Condensed-matter physics/Semiconductors Physical sciences/Materials science/Theory and computation/Computational methods Physical sciences/Mathematics and computing/Computational science Full Text Additional Declarations There is NO Competing Interest. Supplementary Files SupplementaryInformation.pdf Supplementary Information 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-9453126","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Resource","associatedPublications":[],"authors":[{"id":639068325,"identity":"292664d7-8d3e-4389-a80d-cb66d4d1e55c","order_by":0,"name":"Haiping Lu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA30lEQVRIie3RsQrCMBCA4QuFdAm4BgT7Ci0dtFDxVVKcnQRXUwLporj2MQQfwJaAbp0dHOri3NGhg2lHh7SjQ/4x8HEXDsBm+8N8cDg0UIArSg417h+dAYI4yjUht4QDG0sc0hMG48gckJzG7XNGsJPWbAfehJPQN5GIa7KR75BgJHxWQZAXJGTGxYqOcJVcvVTSRAI6AwmLQbJo1f6AUfbRZDWOAFZMLyZBk6Qj5sUUEtFRqkBPEZRVdJ0rvDV+37+L8vFplUew+2qaXbw8ZeJCTeT3BnToKjabzWYb0xdI20YnRbTltAAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-0349-2181","institution":"University of Sheffield","correspondingAuthor":true,"prefix":"","firstName":"Haiping","middleName":"","lastName":"Lu","suffix":""},{"id":639068326,"identity":"4a503dbe-6e90-4dc6-b642-570b68f2d549","order_by":1,"name":"Haolin Wang","email":"","orcid":"","institution":"The University of Sheffield","correspondingAuthor":false,"prefix":"","firstName":"Haolin","middleName":"","lastName":"Wang","suffix":""},{"id":639068327,"identity":"684bf684-9cbb-4a1d-a441-db9477ad6c3e","order_by":2,"name":"Xianyuan Liu","email":"","orcid":"https://orcid.org/0000-0002-3084-519X","institution":"The University of Sheffield","correspondingAuthor":false,"prefix":"","firstName":"Xianyuan","middleName":"","lastName":"Liu","suffix":""},{"id":639068328,"identity":"146bf79c-b340-43bd-8019-b0c3205310a9","order_by":3,"name":"Anna Jungbluth","email":"","orcid":"","institution":"European Space Agency","correspondingAuthor":false,"prefix":"","firstName":"Anna","middleName":"","lastName":"Jungbluth","suffix":""},{"id":639068329,"identity":"fa8ceb1f-5754-4032-a9fb-b1efc031e52f","order_by":4,"name":"Alexandra Ramadan","email":"","orcid":"https://orcid.org/0000-0003-4572-3459","institution":"The University of Sheffield","correspondingAuthor":false,"prefix":"","firstName":"Alexandra","middleName":"","lastName":"Ramadan","suffix":""},{"id":639068330,"identity":"a647b2ce-3e92-4bdd-8946-2b48231bcd01","order_by":5,"name":"Robert Oliver","email":"","orcid":"https://orcid.org/0000-0003-4980-7940","institution":"The University of Sheffield","correspondingAuthor":false,"prefix":"","firstName":"Robert","middleName":"","lastName":"Oliver","suffix":""}],"badges":[],"createdAt":"2026-04-17 23:45:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9453126/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9453126/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109204828,"identity":"58b84244-71df-46fc-95f0-e50a6f5dfb0a","added_by":"auto","created_at":"2026-05-13 15:02:29","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":819010,"visible":true,"origin":"","legend":"Article File","description":"","filename":"Articlefile.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9453126/v1_covered_b1f0a494-0fe9-4d17-bf31-bfa9a8f04c64.pdf"},{"id":109153545,"identity":"281f2e85-da5a-4fd3-8bfc-05bd099ce2ca","added_by":"auto","created_at":"2026-05-13 06:19:00","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":3927766,"visible":true,"origin":"","legend":"Supplementary Information","description":"","filename":"SupplementaryInformation.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9453126/v1/b07e65334f6bf9696d885d2c.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Benchmarking bandgap prediction in semiconductors under experimental and realistic evaluation settings","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":"[email protected]","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-9453126/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9453126/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Accurate bandgap prediction is crucial for semiconductor applications, yet machine learning models trained on computational data often struggle to generalize to experimental bandgap measurements. Challenges related to data fidelity, domain generalization, and model interpretability remain insufficiently addressed in existing evaluation frameworks. To bridge this gap, we introduce RealMat-BaG, a benchmark for assessing model reliability under experimentally relevant conditions. We curate an open-access dataset of experimental bandgaps with aligned crystal structures and compare graph neural networks as well as classical machine learning baselines. Our framework evaluates performance across statistical and domain-based splits, examines transfer from DFT-computed to experimental bandgaps, and analyzes interpretability at both elemental-property and structural levels. Our results reveal the fundamental generalization limitations of current bandgap prediction models and establish a benchmark aligned with experimental measurements for developing more reliable learning strategies for materials discovery.","manuscriptTitle":"Benchmarking bandgap prediction in semiconductors under experimental and realistic evaluation settings","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-05-13 06:18:56","doi":"10.21203/rs.3.rs-9453126/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-computational-science","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"natcomputsci","sideBox":"Learn more about [Nature Computational Science](http://www.nature.com/natcomputsci/)","snPcode":"","submissionUrl":"","title":"Nature Computational Science","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Research","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"52bb0ffd-680b-4ce3-82ee-fd2865bfbc3d","owner":[],"postedDate":"May 13th, 2026","published":true,"recentEditorialEvents":[{"type":"reviewerAgreed","content":"This content is not available.","date":"2026-05-13T03:42:17+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewersInvited","content":"3","date":"2026-05-12T13:41:08+00:00","index":"","fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":68018936,"name":"Physical sciences/Mathematics and computing/Computer science"},{"id":68018937,"name":"Physical sciences/Materials science/Condensed-matter physics/Semiconductors"},{"id":68018938,"name":"Physical sciences/Materials science/Theory and computation/Computational methods"},{"id":68018939,"name":"Physical sciences/Mathematics and computing/Computational science"}],"tags":[],"updatedAt":"2026-05-13T06:18:56+00:00","versionOfRecord":[],"versionCreatedAt":"2026-05-13 06:18:56","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9453126","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9453126","identity":"rs-9453126","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}