ReFair: A Framework for Retention Depolarization in Recommender Systems

ReFair: A Framework for Retention Depolarization in Recommender Systems | 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 ReFair: A Framework for Retention Depolarization in Recommender Systems Junaid ahmed iqbal This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7994140/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 Retention disparities among user groups in recommender systems exacerbate polarization, undermining long-term user engagement and diversity. Addressing this challenge, we propose ReFair, a novel computational framework designed to enhance recommendation algorithms while enforcing retention fairness across the user population. ReFair employs a model-based reinforcement learning approach, iteratively alternating between environment learning for user retention dynamics and fairness-constrained policy optimization. Our method incorporates uncertainties in dynamic estimations to ensure robust performance. Empirical evaluations on real-world datasets demonstrate ReFair's ability to improve long-term user satisfaction and reduce retention polarization effectively. This work introduces new possibilities for fairness in recommender systems, advancing both theoretical and practical objectives. Full Text Additional Declarations No competing interests reported. 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-7994140","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":549979764,"identity":"1a95f04f-257f-4e0b-bc8e-c4f9f4055aa0","order_by":0,"name":"Junaid ahmed iqbal","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA7klEQVRIiWNgGAWjYFCCBIYDDAwSDAYMzI0PgFwePqK0HABrYWw2AGlhI0YLyBqQljYJEJ+gFv727MTDH/5YyG5nP9hW+TXHToaNgfnhoxt4tEicebvhwME2CeOdPYltt2W3JQMdxmZsnIPPmhu5QC0NEokbDgC1SG5jBmrhYZPGp0UepOXAH6CW8w/biiW31RPWYgDWwgbUciOxjfHjtsOEtRiC/HIW5JcZD5ulGbcd52FjJuAXueO5mz9U/KmT3c6ffPDjz23V9vzszQ8f4/U+FDA2AAlmHhCTmQjlcC2MP4hUPQpGwSgYBSMLAACguFN6l5qdTwAAAABJRU5ErkJggg==","orcid":"","institution":"","correspondingAuthor":true,"prefix":"","firstName":"Junaid","middleName":"ahmed","lastName":"iqbal","suffix":""}],"badges":[],"createdAt":"2025-10-31 03:53:30","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7994140/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7994140/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96965435,"identity":"bd8c3e34-891b-425b-bb4c-215fde30f529","added_by":"auto","created_at":"2025-11-28 06:25:25","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3008,"visible":true,"origin":"","legend":"","description":"","filename":"841fb99f34e74e8fbaac4019936fa2fe.json","url":"https://assets-eu.researchsquare.com/files/rs-7994140/v1/aef015b41bf0d3cc5fd8d2af.json"},{"id":105888614,"identity":"f1997bc0-d9e4-4128-8d88-3f8fa2e0207b","added_by":"auto","created_at":"2026-04-01 07:44:40","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":323610,"visible":true,"origin":"","legend":"","description":"","filename":"ReFairAFrameworkforRetentionDepolarizationinRecommenderSystems.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7994140/v1_covered_615f9472-ac47-46e6-93ba-35d422601bdf.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"ReFair: A Framework for Retention Depolarization in Recommender Systems","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"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-7994140/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7994140/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eRetention disparities among user groups in recommender systems exacerbate polarization, undermining long-term user engagement and diversity. Addressing this challenge, we propose ReFair, a novel computational framework designed to enhance recommendation algorithms while enforcing retention fairness across the user population. ReFair employs a model-based reinforcement learning approach, iteratively alternating between environment learning for user retention dynamics and fairness-constrained policy optimization. Our method incorporates uncertainties in dynamic estimations to ensure robust performance. Empirical evaluations on real-world datasets demonstrate ReFair's ability to improve long-term user satisfaction and reduce retention polarization effectively. This work introduces new possibilities for fairness in recommender systems, advancing both theoretical and practical objectives.\u003c/p\u003e","manuscriptTitle":"ReFair: A Framework for Retention Depolarization in Recommender Systems","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-28 06:25:21","doi":"10.21203/rs.3.rs-7994140/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":"52a5d9c9-64d8-4a5b-8eab-60dc5bfe9da0","owner":[],"postedDate":"November 28th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-01T07:43:56+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-28 06:25:21","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7994140","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7994140","identity":"rs-7994140","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}