Modeling Anisotropic Preference Manifolds for Robust Graph-based Fashion Recommendation

Modeling Anisotropic Preference Manifolds for Robust Graph-based Fashion Recommendation | 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 Modeling Anisotropic Preference Manifolds for Robust Graph-based Fashion Recommendation Zhang Xiao, Tan Tien Ping, Zhang Haiping This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8727877/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 10 You are reading this latest preprint version Abstract Graph Convolutional Networks (GNNs) have established themselves as the leading paradigm for collaborative filtering; however, their efficacy in complex domains like fashion is often hindered by a fundamental geometric mismatch between model assumptions and data reality. Conventional methods typically rely on scalar aggregation and Euclidean metrics, which implicitly assume that user interest clusters are isotropic (spherical) and uniformly dense. This assumption fails to capture the complexity of fashion preferences, where style distributions exhibit significant anisotropic variance—ranging from sparse, broad categories to dense, niche trends. To bridge this gap, we propose the Multi-Interest Mahalanobis Denoising Graph Convolutional Network (MIMD-GCN), a framework that synergizes structural disentanglement with geometry-aware denoising. We introduce a Poly-Attention mechanism to disentangle user representations into multiple latent interest centers, thereby resolving the collapse of diverse preferences into single vectors. Furthermore, we construct an anisotropic denoising module based on a learnable Mahalanobis distance barrier. Unlike static Euclidean thresholds, this mechanism dynamically adapts to the covariance structure of specific interest manifolds, establishing elliptical boundaries that effectively isolate spurious interactions while preserving valid niche signals. Extensive experiments on Amazon Fashion and Taobao datasets demonstrate that MIMD-GCN achieves statistically significant improvements over state-of-the-art baselines and maintains high robustness under severe synthetic noise, validating the geometric necessity of non-Euclidean modeling in recommender systems. Physical sciences/Mathematics and computing Physical sciences/Physics Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 31 Mar, 2026 Reviews received at journal 23 Mar, 2026 Reviews received at journal 18 Mar, 2026 Reviewers agreed at journal 23 Feb, 2026 Reviewers agreed at journal 23 Feb, 2026 Reviewers invited by journal 23 Feb, 2026 Editor invited by journal 03 Feb, 2026 Editor assigned by journal 01 Feb, 2026 Submission checks completed at journal 01 Feb, 2026 First submitted to journal 29 Jan, 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-8727877","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":596360252,"identity":"a4f58491-618a-47de-b32d-3ebd128696bd","order_by":0,"name":"Zhang Xiao","email":"","orcid":"","institution":"Hangzhou Dianzi University Information Engineering School","correspondingAuthor":false,"prefix":"","firstName":"Zhang","middleName":"","lastName":"Xiao","suffix":""},{"id":596360253,"identity":"4b678997-5f47-42e9-882f-d40620406bce","order_by":1,"name":"Tan Tien Ping","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAsElEQVRIiWNgGAWjYDACZhBhwMDAT7QOHpgWyQaitcAYBgeI1WLPznzscUHBHbvNN9KfbmDMOUyMw9jSjWcYPEvediPH7AbjNqK08JhJ8xgcTja7kcNGrBb+b2AtxjPSnxFtCxtIi52BRAKxDjvMZiY9w+BwgsSZN2Y3ErelE9bC3n/4mXTBn8P2/O1Ah33cZk1YCwiAYjOxAcRKYGgmXos9lF1HnJZRMApGwSgYUQAAOjQ2mTaE8KUAAAAASUVORK5CYII=","orcid":"","institution":"Universiti Sains Malaysia","correspondingAuthor":true,"prefix":"","firstName":"Tan","middleName":"Tien","lastName":"Ping","suffix":""},{"id":596360254,"identity":"571f0a08-7d2b-4b27-8be0-7284205b974d","order_by":2,"name":"Zhang Haiping","email":"","orcid":"","institution":"Hangzhou Dianzi University Information Engineering School","correspondingAuthor":false,"prefix":"","firstName":"Zhang","middleName":"","lastName":"Haiping","suffix":""}],"badges":[],"createdAt":"2026-01-29 06:38:52","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8727877/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8727877/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104400014,"identity":"d3ce967a-b8c5-43e6-a44e-59232753bde3","added_by":"auto","created_at":"2026-03-11 12:08:31","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":649669,"visible":true,"origin":"","legend":"","description":"","filename":"MIMDGCN1.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8727877/v1_covered_99c871ee-43ed-413b-aae6-700c6a9eb5c5.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Modeling Anisotropic Preference Manifolds for Robust Graph-based Fashion Recommendation","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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8727877/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8727877/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Graph Convolutional Networks (GNNs) have established themselves as the leading paradigm for collaborative filtering; however, their efficacy in complex domains like fashion is often hindered by a fundamental geometric mismatch between model assumptions and data reality. Conventional methods typically rely on scalar aggregation and Euclidean metrics, which implicitly assume that user interest clusters are isotropic (spherical) and uniformly dense. This assumption fails to capture the complexity of fashion preferences, where style distributions exhibit significant anisotropic variance—ranging from sparse, broad categories to dense, niche trends. To bridge this gap, we propose the Multi-Interest Mahalanobis Denoising Graph Convolutional Network (MIMD-GCN), a framework that synergizes structural disentanglement with geometry-aware denoising. We introduce a Poly-Attention mechanism to disentangle user representations into multiple latent interest centers, thereby resolving the collapse of diverse preferences into single vectors. Furthermore, we construct an anisotropic denoising module based on a learnable Mahalanobis distance barrier. Unlike static Euclidean thresholds, this mechanism dynamically adapts to the covariance structure of specific interest manifolds, establishing elliptical boundaries that effectively isolate spurious interactions while preserving valid niche signals. Extensive experiments on Amazon Fashion and Taobao datasets demonstrate that MIMD-GCN achieves statistically significant improvements over state-of-the-art baselines and maintains high robustness under severe synthetic noise, validating the geometric necessity of non-Euclidean modeling in recommender systems.","manuscriptTitle":"Modeling Anisotropic Preference Manifolds for Robust Graph-based Fashion Recommendation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-02 05:41:53","doi":"10.21203/rs.3.rs-8727877/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-04-01T03:39:54+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-24T03:56:02+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-18T22:06:52+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"56686400682257301992975546492763941281","date":"2026-02-23T16:39:20+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"181774216258844856068990075705375030631","date":"2026-02-23T13:49:43+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-23T13:40:43+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-02-03T13:45:46+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-02T03:01:01+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-02T02:59:42+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-01-29T06:31:55+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"f4644d6b-8038-4ccb-8dc3-b83043ef4aaf","owner":[],"postedDate":"March 2nd, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":63455962,"name":"Physical sciences/Mathematics and computing"},{"id":63455963,"name":"Physical sciences/Physics"}],"tags":[],"updatedAt":"2026-04-27T10:39:51+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-02 05:41:53","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8727877","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8727877","identity":"rs-8727877","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}