SILTD: Structural Information for LLM-generated Text Detection

SILTD: Structural Information for LLM-generated Text Detection | 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 SILTD: Structural Information for LLM-generated Text Detection Jing Yang, Shi Wang, Kangli Zi, Yanshun Sun, Yuwei Huang, Tianyu Luo This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4910736/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 23 May, 2025 Read the published version in International Journal of Machine Learning and Cybernetics → Version 1 posted 14 You are reading this latest preprint version Abstract The rapid development of large language models (LLMs) has significantly improved the quality and diversity of AI-generated content(AIGC). LLMs-generated text detection plays an important role in preventing the harmful misuse of large language models. Although many detection methods have been proposed, they all have room for improvement and lack generality when detecting multiple models. To tackle these drawbacks, an unsupervised-based Structural Information for LLM-generated Text Detection(SILTD) method is proposed. First, we construct a multi-relational text graph based on the similarity of text features, which aims to model the intricate similarities and correlations between texts in the generative statistical space. Second, we propose a novel unsupervised graph clustering method. The multi-relational graph is transformed into an encoding tree, which is then optimized based on a two-dimensional structure entropy minimization algorithm to achieve hierarchical clustering of texts. Structural entropy minimization enables achieving high-quality clusters, by measuring the uncertainty of random walks within the graph. Finally, we introduce a new method that measures text similarity and computes the intensity of text aggregation within each cluster, to perform in-cluster label inference. Extensive experiments show that, compared to baseline methods, our approach is more effective and generalizable in detecting six popular LLMs across three datasets. LLMs-generated text detection Structural information Multi-relational graph Clustering Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 23 May, 2025 Read the published version in International Journal of Machine Learning and Cybernetics → Version 1 posted Editorial decision: Revision requested 01 Oct, 2024 Reviews received at journal 24 Sep, 2024 Reviews received at journal 24 Sep, 2024 Reviewers agreed at journal 03 Sep, 2024 Reviews received at journal 28 Aug, 2024 Reviewers agreed at journal 25 Aug, 2024 Reviewers agreed at journal 25 Aug, 2024 Reviewers agreed at journal 25 Aug, 2024 Reviewers agreed at journal 23 Aug, 2024 Reviewers agreed at journal 23 Aug, 2024 Reviewers invited by journal 23 Aug, 2024 Editor assigned by journal 17 Aug, 2024 Submission checks completed at journal 17 Aug, 2024 First submitted to journal 14 Aug, 2024 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-4910736","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":351084535,"identity":"43f141f2-4c15-4fed-b4c9-93241e7c1ee4","order_by":0,"name":"Jing Yang","email":"","orcid":"","institution":"Institute of Computing Technology","correspondingAuthor":false,"prefix":"","firstName":"Jing","middleName":"","lastName":"Yang","suffix":""},{"id":351084537,"identity":"69ac8e03-69f9-4915-b105-fb093d0a7295","order_by":1,"name":"Shi Wang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAw0lEQVRIiWNgGAWjYLACxgYJOTaStRizMTCTpoUhsYFoLQbHzx5++XOHRXof+/kDDD9qGOTNCWo5k5dmzXtGIreNJ5mBsecYg+HOBgJazA7kmBkztgG1SAAdxtvAkGBwgJCW82/MDH+2SaSzAbUw/iVKy40c4we8bRIJIC3MRNlif+ONGTNQiyHQLwaHZY5JGG4gpEWyP8f448+2Onn59oMPH76psZEnaAsQsEnAWEDFEngUIgDzB6KUjYJRMApGwcgFAEsdOah6BJaDAAAAAElFTkSuQmCC","orcid":"","institution":"Institute of Computing Technology","correspondingAuthor":true,"prefix":"","firstName":"Shi","middleName":"","lastName":"Wang","suffix":""},{"id":351084538,"identity":"1fceaac0-b52b-4172-adbd-e0354e332af0","order_by":2,"name":"Kangli Zi","email":"","orcid":"","institution":"Institute of Computing Technology","correspondingAuthor":false,"prefix":"","firstName":"Kangli","middleName":"","lastName":"Zi","suffix":""},{"id":351084539,"identity":"ea8b4ce2-81d3-41e6-8295-4f40a68b55b4","order_by":3,"name":"Yanshun Sun","email":"","orcid":"","institution":"Institute of Computing Technology","correspondingAuthor":false,"prefix":"","firstName":"Yanshun","middleName":"","lastName":"Sun","suffix":""},{"id":351084540,"identity":"fd9a0af2-1844-4b44-a474-ff223eb9d1ed","order_by":4,"name":"Yuwei Huang","email":"","orcid":"","institution":"Institute of Computing Technology","correspondingAuthor":false,"prefix":"","firstName":"Yuwei","middleName":"","lastName":"Huang","suffix":""},{"id":351084541,"identity":"20f7ce91-7acf-435e-af20-08fafc06cb66","order_by":5,"name":"Tianyu Luo","email":"","orcid":"","institution":"Institute of Computing Technology","correspondingAuthor":false,"prefix":"","firstName":"Tianyu","middleName":"","lastName":"Luo","suffix":""}],"badges":[],"createdAt":"2024-08-14 04:51:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4910736/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4910736/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s13042-025-02616-x","type":"published","date":"2025-05-23T15:57:25+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":83460008,"identity":"553b833a-37ea-481d-9d1d-08872399e51c","added_by":"auto","created_at":"2025-05-26 16:08:50","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1994362,"visible":true,"origin":"","legend":"","description":"","filename":"SILTDStructuralInformationforLLMgeneratedTextDetetion.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4910736/v1_covered_74d3a54c-8894-47cc-b99e-14399eb3d5f3.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"SILTD: Structural Information for LLM-generated Text Detection","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":"international-journal-of-machine-learning-and-cybernetics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jmlc","sideBox":"Learn more about [International Journal of Machine Learning and Cybernetics](http://actavetscand.biomedcentral.com/)","snPcode":"13042","submissionUrl":"https://submission.nature.com/new-submission/13042/3","title":"International Journal of Machine Learning and Cybernetics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"LLMs-generated text detection, Structural information, Multi-relational graph, Clustering","lastPublishedDoi":"10.21203/rs.3.rs-4910736/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4910736/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"The rapid development of large language models (LLMs) has significantly improved the quality and diversity of AI-generated content(AIGC). LLMs-generated text detection plays an important role in preventing the harmful misuse of large language models. Although many detection methods have been proposed, they all have room for improvement and lack generality when detecting multiple models. To tackle these drawbacks, an unsupervised-based Structural Information for LLM-generated Text Detection(SILTD) method is proposed. First, we construct a multi-relational text graph based on the similarity of text features, which aims to model the intricate similarities and correlations between texts in the generative statistical space. Second, we propose a novel unsupervised graph clustering method. The multi-relational graph is transformed into an encoding tree, which is then optimized based on a two-dimensional structure entropy minimization algorithm to achieve hierarchical clustering of texts. Structural entropy minimization enables achieving high-quality clusters, by measuring the uncertainty of random walks within the graph. Finally, we introduce a new method that measures text similarity and computes the intensity of text aggregation within each cluster, to perform in-cluster label inference. Extensive experiments show that, compared to baseline methods, our approach is more effective and generalizable in detecting six popular LLMs across three datasets.","manuscriptTitle":"SILTD: Structural Information for LLM-generated Text Detection","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-09-13 07:06:58","doi":"10.21203/rs.3.rs-4910736/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-10-01T07:22:54+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-09-25T03:29:44+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-09-25T03:06:58+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"208553105077026599277798141177769175055","date":"2024-09-04T03:06:32+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-08-28T09:05:01+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"331332400841250793412562956282037074546","date":"2024-08-25T14:27:46+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"106656563400618239641414788391514539139","date":"2024-08-25T11:04:56+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"103208176132709675356956661629109767172","date":"2024-08-25T08:56:34+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"240355439384573110458789171223706956168","date":"2024-08-23T12:36:43+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"148898762287421851651306904202847853792","date":"2024-08-23T10:51:59+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-08-23T08:38:34+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-08-18T01:22:23+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-08-17T06:12:02+00:00","index":"","fulltext":""},{"type":"submitted","content":"International Journal of Machine Learning and Cybernetics","date":"2024-08-14T04:49:17+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"international-journal-of-machine-learning-and-cybernetics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jmlc","sideBox":"Learn more about [International Journal of Machine Learning and Cybernetics](http://actavetscand.biomedcentral.com/)","snPcode":"13042","submissionUrl":"https://submission.nature.com/new-submission/13042/3","title":"International Journal of Machine Learning and Cybernetics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"68a40b95-c170-4f53-891c-499c4d1c7fef","owner":[],"postedDate":"September 13th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-05-26T16:00:43+00:00","versionOfRecord":{"articleIdentity":"rs-4910736","link":"https://doi.org/10.1007/s13042-025-02616-x","journal":{"identity":"international-journal-of-machine-learning-and-cybernetics","isVorOnly":false,"title":"International Journal of Machine Learning and Cybernetics"},"publishedOn":"2025-05-23 15:57:25","publishedOnDateReadable":"May 23rd, 2025"},"versionCreatedAt":"2024-09-13 07:06:58","video":"","vorDoi":"10.1007/s13042-025-02616-x","vorDoiUrl":"https://doi.org/10.1007/s13042-025-02616-x","workflowStages":[]},"version":"v1","identity":"rs-4910736","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4910736","identity":"rs-4910736","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}