Detecting Obfuscated Cross-Site Scripting Attacks Using LLM-Generated Payloads and Graph Neural Networks

Detecting Obfuscated Cross-Site Scripting Attacks Using LLM-Generated Payloads and Graph Neural Networks | 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 Detecting Obfuscated Cross-Site Scripting Attacks Using LLM-Generated Payloads and Graph Neural Networks Abdelkader TAJTIT, Mohammed Serrhini This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9254071/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 12 You are reading this latest preprint version Abstract Cross-Site Scripting (XSS) remains a persistent web security vulnerability, particularly in its obfuscated forms. Building upon recent advancements in leveraging Large Language Models (LLMs) for security, this study adopts a novel pipeline that generates complex obfuscated XSS samples using a fine-tuned code generation model \texttt{CodeGen-350M-multi}. These samples are transformed into Control Flow Graphs (CFGs) to capture their structural and behavioral semantics. We then utilize a Graph Neural Network (GNN) architecture to classify these samples. Our approach achieves high performance on held-out test data (reported accuracy: 99.72%, precision: 99.60%, recall: 100%, F1-score: 99.70%). We additionally extend the evaluation to out-of-distribution sources, runtime feasibility and ablation analysis to validate robustness and reproducibility. Cross-Site Scripting (XSS) Large Language Models (LLMs) Code Obfuscation Graph Neural Networks Control Flow Graphs (CFG) Web Security Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 12 May, 2026 Reviews received at journal 26 Apr, 2026 Reviews received at journal 26 Apr, 2026 Reviewers agreed at journal 26 Apr, 2026 Reviews received at journal 22 Apr, 2026 Reviewers agreed at journal 18 Apr, 2026 Reviewers agreed at journal 17 Apr, 2026 Reviewers agreed at journal 17 Apr, 2026 Reviewers invited by journal 17 Apr, 2026 Editor assigned by journal 30 Mar, 2026 Submission checks completed at journal 30 Mar, 2026 First submitted to journal 28 Mar, 2026 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-9254071","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":627977931,"identity":"3f6496c2-267a-464a-956a-b0435d122805","order_by":0,"name":"Abdelkader TAJTIT","email":"data:image/png;base64,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","orcid":"","institution":"Mohamed I University","correspondingAuthor":true,"prefix":"","firstName":"Abdelkader","middleName":"","lastName":"TAJTIT","suffix":""},{"id":627977932,"identity":"4b273d0c-c4bd-41f5-875e-13e58c5c6160","order_by":1,"name":"Mohammed Serrhini","email":"","orcid":"","institution":"Mohamed I University","correspondingAuthor":false,"prefix":"","firstName":"Mohammed","middleName":"","lastName":"Serrhini","suffix":""}],"badges":[],"createdAt":"2026-03-28 16:08:43","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9254071/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9254071/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107708111,"identity":"94ecb6be-17ad-43fd-9abe-4330b6d7e198","added_by":"auto","created_at":"2026-04-24 09:21:58","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":471948,"visible":true,"origin":"","legend":"","description":"","filename":"JournalofComputerVirologyandHackingTechniques.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9254071/v1_covered_4b9537db-58f4-4075-89b9-cf08ba9e23c9.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Detecting Obfuscated Cross-Site Scripting Attacks Using LLM-Generated Payloads and Graph Neural Networks","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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":"journal-of-computer-virology-and-hacking-techniques","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jicv","sideBox":"Learn more about [Journal of Computer Virology and Hacking Techniques](http://link.springer.com/journal/11416)","snPcode":"11416","submissionUrl":"https://submission.springernature.com/new-submission/11416/3","title":"Journal of Computer Virology and Hacking Techniques","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Cross-Site Scripting (XSS), Large Language Models (LLMs), Code Obfuscation, Graph Neural Networks, Control Flow Graphs (CFG), Web Security","lastPublishedDoi":"10.21203/rs.3.rs-9254071/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9254071/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eCross-Site Scripting (XSS) remains a persistent web security vulnerability, particularly in its obfuscated forms. Building upon recent advancements in leveraging Large Language Models (LLMs) for security, this study adopts a novel pipeline that generates complex obfuscated XSS samples using a fine-tuned code generation model \\texttt{CodeGen-350M-multi}. These samples are transformed into Control Flow Graphs (CFGs) to capture their structural and behavioral semantics. We then utilize a Graph Neural Network (GNN) architecture to classify these samples. Our approach achieves high performance on held-out test data (reported accuracy: 99.72%, precision: 99.60%, recall: 100%, F1-score: 99.70%). We additionally extend the evaluation to out-of-distribution sources, runtime feasibility and ablation analysis to validate robustness and reproducibility.\u003c/p\u003e","manuscriptTitle":"Detecting Obfuscated Cross-Site Scripting Attacks Using LLM-Generated Payloads and Graph Neural Networks","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-24 04:48:06","doi":"10.21203/rs.3.rs-9254071/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-13T01:11:53+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-26T15:17:58+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-26T04:52:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"158773247823164871494899312620006066236","date":"2026-04-26T04:23:39+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-22T19:58:04+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"165086076613311812243820997282913208903","date":"2026-04-19T02:27:47+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"226337677974588353016273537745326898401","date":"2026-04-17T05:43:32+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"294567667425648084870017250922682307126","date":"2026-04-17T04:55:36+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-17T04:39:36+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-03-30T13:37:48+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-03-30T13:37:13+00:00","index":"","fulltext":""},{"type":"submitted","content":"Journal of Computer Virology and Hacking Techniques","date":"2026-03-28T16:02:56+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"journal-of-computer-virology-and-hacking-techniques","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jicv","sideBox":"Learn more about [Journal of Computer Virology and Hacking Techniques](http://link.springer.com/journal/11416)","snPcode":"11416","submissionUrl":"https://submission.springernature.com/new-submission/11416/3","title":"Journal of Computer Virology and Hacking Techniques","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"8af60990-0baa-44d9-9fb4-e09303f362dc","owner":[],"postedDate":"April 24th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-13T01:11:53+00:00","index":17,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-24T04:48:06+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-24 04:48:06","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9254071","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9254071","identity":"rs-9254071","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}