DSA-GNAS: Graph Neural Architecture Search with Deep Semantic Adaption of Large Language Models

DSA-GNAS: Graph Neural Architecture Search with Deep Semantic Adaption of Large Language 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 Research Article DSA-GNAS: Graph Neural Architecture Search with Deep Semantic Adaption of Large Language Models Siyang Xiao, Jiamin Chen, Zhenpeng Wu, Shuqing Wu, Jianliang Gao This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6639133/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 31 Mar, 2026 Read the published version in Knowledge and Information Systems → Version 1 posted 11 You are reading this latest preprint version Abstract The remarkable success of pre-trained large language model (LLM) in naturallanguage processing develops a new paradigm of combining LLM with graph neu-ral networks (GNN) on modeling textual-attributed graph. However, manuallydesigning the optimal model architectures to adapt the deep semantic of LLM ondifferent graphs is trivial and highly demands on expertise knowledge. Thoughgraph neural architecture search (GNAS) provide a feasible solutions to auto-matically design optimal GNN architectures for different graphs, former researchmainly establish on the shallow embedding methods, which ignore the differencebetween the deep semantic space and shallow embeddings. Focus on these issues,we propose DSA-GNAS, an effective TAG learning framework with auto graphneural architecture search and deep semantic adaption. To better leverage thedeep semantic space, we propose a novel Structure-Semantic Fusion (S2F) searchspace. The model architectures are sampled from the S2F space and form into adual-path adapter to fine-tune semantic embedding generated by LLM, which cansufficiently make adaptation on the semantic space to graph downstream task.The model architectures are automatically optimized through a genetic searchstrategy, which is global and not restricted by gradient, offering promising effi-ciency in searching for the optimal model architecture. Experimental results showthat DSA-GNAS can significantly improve performance on the graph task overother baselines, demonstrating that DSA-GNAS can effectively work in design-ing optimal model architectures for adapting the deep semantic to graph relatedtasks. Graph neural architecture search Graph neural network Textual attributed graph Large language model Deep semantic Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 31 Mar, 2026 Read the published version in Knowledge and Information Systems → Version 1 posted Editorial decision: Revision requested 07 Aug, 2025 Reviews received at journal 31 Jul, 2025 Reviews received at journal 26 Jul, 2025 Reviewers agreed at journal 20 Jul, 2025 Reviewers agreed at journal 15 Jul, 2025 Reviewers agreed at journal 14 Jul, 2025 Reviewers agreed at journal 11 Jun, 2025 Reviewers invited by journal 11 Jun, 2025 Editor assigned by journal 29 May, 2025 Submission checks completed at journal 16 May, 2025 First submitted to journal 11 May, 2025 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-6639133","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":470622190,"identity":"42981fa5-391b-4a5a-b87b-1c30543a9db2","order_by":0,"name":"Siyang Xiao","email":"","orcid":"","institution":"Central South University","correspondingAuthor":false,"prefix":"","firstName":"Siyang","middleName":"","lastName":"Xiao","suffix":""},{"id":470622191,"identity":"3cebdf5c-7520-4972-a33f-3e90c0d1aa22","order_by":1,"name":"Jiamin Chen","email":"","orcid":"","institution":"Guizhou Normal University","correspondingAuthor":false,"prefix":"","firstName":"Jiamin","middleName":"","lastName":"Chen","suffix":""},{"id":470622192,"identity":"75fd2e8b-7966-4c3e-91f4-a7343108c78a","order_by":2,"name":"Zhenpeng Wu","email":"","orcid":"","institution":"Central South University","correspondingAuthor":false,"prefix":"","firstName":"Zhenpeng","middleName":"","lastName":"Wu","suffix":""},{"id":470622194,"identity":"d01b41ff-8b20-4c77-80af-adcb46d6a0bf","order_by":3,"name":"Shuqing Wu","email":"","orcid":"","institution":"Central South University","correspondingAuthor":false,"prefix":"","firstName":"Shuqing","middleName":"","lastName":"Wu","suffix":""},{"id":470622199,"identity":"c445a0fe-8473-4617-a728-1699e5fb9191","order_by":4,"name":"Jianliang Gao","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABAklEQVRIie2RMUvEMBiGUwrp8qFrBrV/IUegeOju30gQOkURBKdwfCDcJNfVv+HiXAhk0t3hhgOhix06yU1iIufY0PHg8gwvHyEP7xdCSCKxp3BJyAkN07A7mKRAULLnqUoAQuQwRSmvbHu3MWs4Kt7t16VZKCweXxkx61Fl9lZLLl0HFG7rC+2sQnAPjLhuXEHNuaTWv0VX4gZbhUxXLEM7rjS9V368ctxXYo5+sbKPKyXzLWrpFabFZ4a5b4G4wlknuVoFpa+yJ2fFEur7c+kiLc21m22/7VnZaDFszeK0KezLx2AiLS2h/99A2V+GkKOCb0GSb3ZzPkQuJhKJxAHzC2TEUWwvpQ8/AAAAAElFTkSuQmCC","orcid":"","institution":"Central South University","correspondingAuthor":true,"prefix":"","firstName":"Jianliang","middleName":"","lastName":"Gao","suffix":""}],"badges":[],"createdAt":"2025-05-11 10:53:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6639133/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6639133/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s10115-026-02732-2","type":"published","date":"2026-03-31T15:58:19+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":106343421,"identity":"86fefd77-aa5b-4e31-be85-e6c9d7b35394","added_by":"auto","created_at":"2026-04-07 16:05:34","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1026268,"visible":true,"origin":"","legend":"","description":"","filename":"DSAGNASGraphNeuralArchitectureSearchwithDeepSemanticAdaptionofLargeLanguageModelsv2submit.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6639133/v1_covered_6d252520-5381-4b9a-a0eb-8d31ef03622c.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"DSA-GNAS: Graph Neural Architecture Search with Deep Semantic Adaption of Large Language Models","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"knowledge-and-information-systems","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"kais","sideBox":"Learn more about [Knowledge and Information Systems](http://link.springer.com/journal/10115)","snPcode":"10115","submissionUrl":"https://submission.nature.com/new-submission/10115/3","title":"Knowledge and Information Systems","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Graph neural architecture search, Graph neural network, Textual attributed graph, Large language model, Deep semantic","lastPublishedDoi":"10.21203/rs.3.rs-6639133/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6639133/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"The remarkable success of pre-trained large language model (LLM) in naturallanguage processing develops a new paradigm of combining LLM with graph neu-ral networks (GNN) on modeling textual-attributed graph. However, manuallydesigning the optimal model architectures to adapt the deep semantic of LLM ondifferent graphs is trivial and highly demands on expertise knowledge. Thoughgraph neural architecture search (GNAS) provide a feasible solutions to auto-matically design optimal GNN architectures for different graphs, former researchmainly establish on the shallow embedding methods, which ignore the differencebetween the deep semantic space and shallow embeddings. Focus on these issues,we propose DSA-GNAS, an effective TAG learning framework with auto graphneural architecture search and deep semantic adaption. To better leverage thedeep semantic space, we propose a novel Structure-Semantic Fusion (S2F) searchspace. The model architectures are sampled from the S2F space and form into adual-path adapter to fine-tune semantic embedding generated by LLM, which cansufficiently make adaptation on the semantic space to graph downstream task.The model architectures are automatically optimized through a genetic searchstrategy, which is global and not restricted by gradient, offering promising effi-ciency in searching for the optimal model architecture. Experimental results showthat DSA-GNAS can significantly improve performance on the graph task overother baselines, demonstrating that DSA-GNAS can effectively work in design-ing optimal model architectures for adapting the deep semantic to graph relatedtasks.","manuscriptTitle":"DSA-GNAS: Graph Neural Architecture Search with Deep Semantic Adaption of Large Language Models","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-13 06:33:17","doi":"10.21203/rs.3.rs-6639133/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-08-07T09:48:00+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-31T07:46:06+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-07-27T03:27:18+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"218801782168480183715018972016661073960","date":"2025-07-21T00:30:34+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"96263702481892257882544724127172389338","date":"2025-07-15T04:00:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"64572110463450467593329041976056627982","date":"2025-07-14T07:14:19+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"273818742394046859997762897136480840216","date":"2025-06-11T14:29:02+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-06-11T14:07:18+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-05-30T03:31:10+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-05-16T10:45:11+00:00","index":"","fulltext":""},{"type":"submitted","content":"Knowledge and Information Systems","date":"2025-05-11T10:43:17+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"knowledge-and-information-systems","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"kais","sideBox":"Learn more about [Knowledge and Information Systems](http://link.springer.com/journal/10115)","snPcode":"10115","submissionUrl":"https://submission.nature.com/new-submission/10115/3","title":"Knowledge and Information Systems","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"da1ee926-79c4-42ee-8792-68eb4e693ffc","owner":[],"postedDate":"June 13th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-04-07T16:02:43+00:00","versionOfRecord":{"articleIdentity":"rs-6639133","link":"https://doi.org/10.1007/s10115-026-02732-2","journal":{"identity":"knowledge-and-information-systems","isVorOnly":false,"title":"Knowledge and Information Systems"},"publishedOn":"2026-03-31 15:58:19","publishedOnDateReadable":"March 31st, 2026"},"versionCreatedAt":"2025-06-13 06:33:17","video":"","vorDoi":"10.1007/s10115-026-02732-2","vorDoiUrl":"https://doi.org/10.1007/s10115-026-02732-2","workflowStages":[]},"version":"v1","identity":"rs-6639133","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6639133","identity":"rs-6639133","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}