Bi-Predictability: A Real-Time Signal for Monitoring LLM Interaction Integrity

Bi-Predictability: A Real-Time Signal for Monitoring LLM Interaction Integrity | 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 Bi-Predictability: A Real-Time Signal for Monitoring LLM Interaction Integrity Wael Hafez, Amir Nazeri This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9182401/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Large language models (LLMs) are increasingly deployed in multi-turn workflows where reliability depends on maintaining interaction integrity over time. Current evaluation methods are poorly matched to this setting: judge-based systems are post hoc and costly, while token-level measures such as perplexity capture output uncertainty but not whether the interaction remains structurally coupled. Here we show that interaction integrity can be monitored continuously using bi-predictability (𝑃), an information-theoretic measure computed from token-frequency statistics across the context-response-next-prompt loop. We operationalize 𝑃 through the Information Digital Twin (IDT), a lightweight architecture that estimates coupling from the observable token stream alone, without embeddings, auxiliary evaluators, or access to model internals. Across 4,500 turns between one student model and three frontier teacher models, the IDT detected all tested perturbations, including contradictions, topic shifts, and non-sequiturs, with 100% sensitivity, matching costlier methods at a fraction of the overhead. Structural coupling and semantic quality proved empirically separable: 𝑃 aligned with structural consistency in 85% of conditions but with semantic scores in only 44%, revealing a regime of silent uncoupling in which responses remain strong while interaction integrity degrades. These results establish 𝑃 as a practical, low-cost, real-time drift monitoring signal and suggest that structural and semantic evaluation should serve as complementary layers in reliable LLM deployment Artificial Intelligence and Machine Learning Large language models Multi-turn interaction Real-time monitoring Interaction integrity Full Text Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted 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-9182401","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":609696820,"identity":"26945223-0ba6-4f51-ac0c-b04814e0962c","order_by":0,"name":"Wael Hafez","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAwklEQVRIiWNgGAWjYFACHgaGCgYGORK1nGFgMCZdS2ID0RoMDvAe/HCgoi59fnt34ufKNgZ5frEDhLTwJUscOMOWu+HM2c2SZ84wGM6cnYBfi9kBHgPpj208uRskcjdINlQwJBjcJqzF+MfBfxLp8vPfbv7ZYECcFjOJgw0GCQw3eLcRZ4v9YR4ziwPHEgw3nMndZtlwRoKwXyTbe4xvHKipk5dvP7v5ZmObjTy/NAEtDMyoXAkCykfBKBgFo2AUEAUAep5DoeZOQPYAAAAASUVORK5CYII=","orcid":"","institution":"Semarx Research LLC","correspondingAuthor":true,"prefix":"","firstName":"Wael","middleName":"","lastName":"Hafez","suffix":""},{"id":609696821,"identity":"fff74ebf-4f25-4f0d-b663-52e3939f9027","order_by":1,"name":"Amir Nazeri","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2UlEQVRIiWNgGAWjYFACxgYYo/EByVqaDUi2j02CKGX80w63fWDcYWfPL324rfJHxbZ8/hkJjB8+5uDWInE7sXkG45lkZsm+xLbbPGduW844c4BZcuY2PNYAtTAwtjGzGZxhbLsNRAYMxxvYmHnxaJGHaKnnsQdqKfz577aB/GEG/FoMIFoOSxjwMLYx8DbcNjAgZIshSEti23EDiTOMzdI8x24bGJ452IzXL3K30x8zfGyrtufvYX/48UfNbQO5G8kHP3zE530QSEDlwiN3FIyCUTAKRgG5AACHK05bZ331rAAAAABJRU5ErkJggg==","orcid":"","institution":"Semarx Research LLC","correspondingAuthor":true,"prefix":"","firstName":"Amir","middleName":"","lastName":"Nazeri","suffix":""}],"badges":[],"createdAt":"2026-03-21 00:45:32","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-9182401/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9182401/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105564738,"identity":"c1e95b5a-28c3-40ba-a00a-7486f00a5dbb","added_by":"auto","created_at":"2026-03-27 12:50:42","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":506092,"visible":true,"origin":"","legend":"","description":"","filename":"HafezLLMcom.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9182401/v1_covered_b7cad722-31d8-4f10-a0d4-f9f7f2e33ee9.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eBi-Predictability: A Real-Time Signal for Monitoring LLM Interaction Integrity\u003c/strong\u003e\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Semarx Research LLC","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Large language models, Multi-turn interaction, Real-time monitoring, Interaction integrity","lastPublishedDoi":"10.21203/rs.3.rs-9182401/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9182401/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eLarge language models (LLMs) are increasingly deployed in multi-turn workflows where reliability depends on maintaining interaction integrity over time. Current evaluation methods are poorly matched to this setting: judge-based systems are post hoc and costly, while token-level measures such as perplexity capture output uncertainty but not whether the interaction remains structurally coupled. Here we show that interaction integrity can be monitored continuously using bi-predictability (𝑃), an information-theoretic measure computed from token-frequency statistics across the context-response-next-prompt loop. We operationalize 𝑃 through the Information Digital Twin (IDT), a lightweight architecture that estimates coupling from the observable token stream alone, without embeddings, auxiliary evaluators, or access to model internals. Across 4,500 turns between one student model and three frontier teacher models, the IDT detected all tested perturbations, including contradictions, topic shifts, and non-sequiturs, with 100% sensitivity, matching costlier methods at a fraction of the overhead. Structural coupling and semantic quality proved empirically separable: 𝑃 aligned with structural consistency in 85% of conditions but with semantic scores in only 44%, revealing a regime of silent uncoupling in which responses remain strong while interaction integrity degrades. These results establish 𝑃 as a practical, low-cost, real-time drift monitoring signal and suggest that structural and semantic evaluation should serve as complementary layers in reliable LLM deployment\u003c/p\u003e","manuscriptTitle":"Bi-Predictability: A Real-Time Signal for Monitoring LLM Interaction Integrity","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-24 04:11:25","doi":"10.21203/rs.3.rs-9182401/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"4125f34d-cd79-4118-9ce0-343e6924e357","owner":[],"postedDate":"March 24th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":64879147,"name":"Artificial Intelligence and Machine Learning"}],"tags":[],"updatedAt":"2026-03-24T04:11:25+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-24 04:11:25","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9182401","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9182401","identity":"rs-9182401","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}