AionRAG: Time-Correct Retrieval-Augmented Generation under Knowledge Drift

AionRAG: Time-Correct Retrieval-Augmented Generation under Knowledge Drift | 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 Article AionRAG: Time-Correct Retrieval-Augmented Generation under Knowledge Drift Rui Li, Shuang Cao, Ruihua Liu, Alexandre Duprey, Angel Dong This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8912660/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 Retrieval-augmented generation (RAG) can fail in dynamic corpora due to time incorrectness: semantically relevant retrieval mixes multiple historical versions of the same claim, and the model often fails to resolve contradictions in favor of the version that is valid at the query time. We introduce AionRAG, a time-correct RAG system that treats retrieval as a calibrated control problem. Given a query, AionRAG predicts whether retrieval is needed and, when it is, selects a query-specific evidence window and hop depth; it then filters candidates by time before semantic ranking to prevent version mixing, and applies a lightweight conflict-gated decoding rule when retrieved evidence disagrees with the model prior. Crucially, AionRAG calibrates decision confidence (ECE=1.7%) so a single threshold maps to predictable latency–quality trade-offs across domains. Across seven benchmarks (242,900 queries) spanning controlled drift tests and real-world evolving corpora (Wikipedia revision histories, U.S. Federal Register policies, and licensed financial news), AionRAG improves temporal consistency and faithfulness while reducing retrieval calls by 29%. On WikiRevision-Real, AionRAG improves temporal consistency by +7.2 points [95% CI: 6.1, 8.3] and faithfulness by +6.4 points [5.3, 7.5] over RouterEns+NLI (our strongest deployable baseline). Long-context baselines (128k tokens) remain vulnerable to conflict amplification, trailing AionRAG by 12.5 points on high-conflict queries despite 2.9x higher latency. These results position time-correct retrieval control as a first-class requirement for reliable RAG under knowledge drift. Physical sciences/Mathematics and computing/Computer science Scientific community and society/Business and industry/Technology Full Text Additional Declarations There is NO Competing Interest. Supplementary Files nmisupplementary.pdf Supplementary File 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-8912660","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":593951726,"identity":"e94dabbd-d8d5-40eb-8060-39a7feeb0251","order_by":0,"name":"Rui Li","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA20lEQVRIiWNgGAWjYFACxgaGBwY2CWzszA1QER4gNiCgJcEgLYGNmZFoLUCQwHA4gQFVCx5gPiMZaEvB+Tw+Zsa2Dx/+2Mibt589wPCjALcWmRuJIIfdLgY6rHnmzLY0wzln8hIYe/A4TEICoiWxDaiFmbfhMOMMCR4DBh7CWs5BtPz5898epIXxD2EtByBaGNgOJIK0MOO1hedhw4EEg2SwFsbetuTkGTw5Bodl8GlhT3/44MMfu8T57c2HGX78sbOdwX7G8OGbP7i1gMABIkRGwSgYBaNgFJAEAJ6MSocllulrAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0009-0007-6857-2127","institution":"Hill Research","correspondingAuthor":true,"prefix":"","firstName":"Rui","middleName":"","lastName":"Li","suffix":""},{"id":593951727,"identity":"1b641701-ad12-4b2c-9cea-511944d13b75","order_by":1,"name":"Shuang Cao","email":"","orcid":"","institution":"Hill Research","correspondingAuthor":false,"prefix":"","firstName":"Shuang","middleName":"","lastName":"Cao","suffix":""},{"id":593951728,"identity":"dd7ad318-d65f-4924-a1b2-18cd05658896","order_by":2,"name":"Ruihua Liu","email":"","orcid":"","institution":"Hill Research","correspondingAuthor":false,"prefix":"","firstName":"Ruihua","middleName":"","lastName":"Liu","suffix":""},{"id":593951729,"identity":"f13b9441-ab32-4048-8abe-37dbdf1786b5","order_by":3,"name":"Alexandre Duprey","email":"","orcid":"","institution":"Hill Research","correspondingAuthor":false,"prefix":"","firstName":"Alexandre","middleName":"","lastName":"Duprey","suffix":""},{"id":593951730,"identity":"9ae7c626-fdfc-41c1-ae42-6e4d239f9914","order_by":4,"name":"Angel Dong","email":"","orcid":"","institution":"Hill Research","correspondingAuthor":false,"prefix":"","firstName":"Angel","middleName":"","lastName":"Dong","suffix":""}],"badges":[],"createdAt":"2026-02-18 22:45:27","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8912660/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8912660/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":103037972,"identity":"7681a479-bd01-4fea-9d41-da18565e1dd3","added_by":"auto","created_at":"2026-02-20 02:53:03","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":380351,"visible":true,"origin":"","legend":"Article File","description":"","filename":"nmi.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8912660/v1_covered_c76ee860-4211-499f-b4d3-98b5b541cfc7.pdf"},{"id":103037971,"identity":"5656fe78-c0c2-4bc8-aaeb-e8acf36a8fc2","added_by":"auto","created_at":"2026-02-20 02:52:58","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":604320,"visible":true,"origin":"","legend":"Supplementary File","description":"","filename":"nmisupplementary.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8912660/v1/39da6d60816ba9c5da869827.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"AionRAG: Time-Correct Retrieval-Augmented Generation under Knowledge Drift","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"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":"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":"","lastPublishedDoi":"10.21203/rs.3.rs-8912660/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8912660/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Retrieval-augmented generation (RAG) can fail in dynamic corpora due to time incorrectness: semantically relevant retrieval mixes multiple historical versions of the same claim, and the model often fails to resolve contradictions in favor of the version that is valid at the query time. We introduce AionRAG, a time-correct RAG system that treats retrieval as a calibrated control problem. Given a query, AionRAG predicts whether retrieval is needed and, when it is, selects a query-specific evidence window and hop depth; it then filters candidates by time before semantic ranking to prevent version mixing, and applies a lightweight conflict-gated decoding rule when retrieved evidence disagrees with the model prior. Crucially, AionRAG calibrates decision confidence (ECE=1.7%) so a single threshold maps to predictable latency–quality trade-offs across domains. Across seven benchmarks (242,900 queries) spanning controlled drift tests and real-world evolving corpora (Wikipedia revision histories, U.S. Federal Register policies, and licensed financial news), AionRAG improves temporal consistency and faithfulness while reducing retrieval calls by 29%. On WikiRevision-Real, AionRAG improves temporal consistency by +7.2 points [95% CI: 6.1, 8.3] and faithfulness by +6.4 points [5.3, 7.5] over RouterEns+NLI (our strongest deployable baseline). Long-context baselines (128k tokens) remain vulnerable to conflict amplification, trailing AionRAG by 12.5 points on high-conflict queries despite 2.9x higher latency. These results position time-correct retrieval control as a first-class requirement for reliable RAG under knowledge drift.","manuscriptTitle":"AionRAG: Time-Correct Retrieval-Augmented Generation under Knowledge Drift","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-20 02:52:53","doi":"10.21203/rs.3.rs-8912660/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":"3db85637-f94e-4c82-aaad-520d73c18fe1","owner":[],"postedDate":"February 20th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":63205537,"name":"Physical sciences/Mathematics and computing/Computer science"},{"id":63205538,"name":"Scientific community and society/Business and industry/Technology"}],"tags":[],"updatedAt":"2026-02-20T02:52:53+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-20 02:52:53","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8912660","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8912660","identity":"rs-8912660","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}