Cross-Modal Local Interest Contrast with Dual-Graph Denoising for Multimodal Recommendation

Cross-Modal Local Interest Contrast with Dual-Graph Denoising for Multimodal 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 Research Article Cross-Modal Local Interest Contrast with Dual-Graph Denoising for Multimodal Recommendation Yuxin Qi, Quangui Zhang, Xinqiang Ma, Xie Feng, Qiang Li, Yi Huang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7494692/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 16 Jan, 2026 Read the published version in Signal, Image and Video Processing → Version 1 posted 4 You are reading this latest preprint version Abstract Multimodal recommender systems improve recommendation accuracy by incorporating item multimodal features (e.g., text, images) alongside user-item interactions. However, they face two critical challenges: (1) local user interests in multimodal features are often obscured by irrelevant content (e.g., background clutter in product images), and (2) behavioral data contains pervasive low-credibility interactions (e.g., accidental clicks) that propagate noise through graph-based recommenders. Notably, over-reliance on region-of-interest (ROI) features during graph construction may introduce spurious edges by ignoring global contextual relationships, exacerbating semantic distortion. To address these issues, we propose CLID (Cross-modal Local Interest Denoising), a novel framework integrating C ross-Modal L ocal I nterest Contrast and Dual-Graph D enoising. First, our local interest contrast mechanism employs text-guided visual attention alignment and a contrastive loss function to enhance discriminative local features—for example, it learns to focus on "sleeve design" in clothing images while suppressing unrelated background features. Crucially, it adaptively weights local features against global representations to prevent ROI-induced bias. Second, the dual-graph denoising architecture combines: (i) a local graph that stabilizes neighbor aggregation via structural consistency to attenuate noisy interactions and (ii) a hypergraph capturing group-wise behavioral patterns, where high-confidence interactions are reinforced through co-occurrence frequency weighting. Experiments demonstrate that CLID significantly improves recommendation performance on three Amazon review datasets: Baby; Clothing, Shoes and Jewelry; as well as Sports and Outdoors. The proposed CLID framework provides a generalizable contrast-and-denoise paradigm for robust multimodal recommendation, effectively bridging fine-grained feature enhancement with noise-resilient graph learning.The code implementation is openly available at the following repository: https://github.com/Qiyx5025/CLID-master. Multimodal recommendation Cross-modal alignment Contrastive learning Hypergraph learning Graph denoising. Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 16 Jan, 2026 Read the published version in Signal, Image and Video Processing → Version 1 posted Editorial decision: Revision requested 11 Sep, 2025 Editor assigned by journal 02 Sep, 2025 Submission checks completed at journal 02 Sep, 2025 First submitted to journal 30 Aug, 2025 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-7494692","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":513844452,"identity":"185a83fb-00d3-420d-a9ba-78a7ac8cbe93","order_by":0,"name":"Yuxin Qi","email":"","orcid":"","institution":"Liaoning Technical University","correspondingAuthor":false,"prefix":"","firstName":"Yuxin","middleName":"","lastName":"Qi","suffix":""},{"id":513844453,"identity":"19acfce0-643f-4216-95e2-ce113c13c5c2","order_by":1,"name":"Quangui Zhang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAs0lEQVRIiWNgGAWjYDACCRBhYMPDz95AmpY0GcmeAyRpYThsY3DDgUgd8rObN7/mKTjPw3CDgfHDxxwitDDOOVZmzWNwm4dxdgOz5MxtRGhhlsgxMwZpYZY5wMbMS4wWNoiWczxsEglEauGRyDF+zGNwgIeHaC0SEmlljHMMknkkeA42E+cX+RnJmz+8+WNnb3+8+eCHj8RoAQIzSNQwMDYQpx4IjD8QrXQUjIJRMApGJgAALWwu51g3VkwAAAAASUVORK5CYII=","orcid":"","institution":"Chongqing University of Arts and Sciences","correspondingAuthor":true,"prefix":"","firstName":"Quangui","middleName":"","lastName":"Zhang","suffix":""},{"id":513844454,"identity":"ae8c137d-a850-40fb-8239-496968fd703d","order_by":2,"name":"Xinqiang Ma","email":"","orcid":"","institution":"Chongqing University of Arts and Sciences","correspondingAuthor":false,"prefix":"","firstName":"Xinqiang","middleName":"","lastName":"Ma","suffix":""},{"id":513844455,"identity":"0ded0f01-1548-4769-8569-341b92592230","order_by":3,"name":"Xie Feng","email":"","orcid":"","institution":"Chongqing University of Arts and Sciences","correspondingAuthor":false,"prefix":"","firstName":"Xie","middleName":"","lastName":"Feng","suffix":""},{"id":513844456,"identity":"fc6d472d-4c47-47de-8719-1cc0fbb86f69","order_by":4,"name":"Qiang Li","email":"","orcid":"","institution":"Chongqing University of Arts and Sciences","correspondingAuthor":false,"prefix":"","firstName":"Qiang","middleName":"","lastName":"Li","suffix":""},{"id":513844458,"identity":"3a64db80-1173-4066-bea5-b39ec3c80245","order_by":5,"name":"Yi Huang","email":"","orcid":"","institution":"Chongqing University of Arts and Sciences","correspondingAuthor":false,"prefix":"","firstName":"Yi","middleName":"","lastName":"Huang","suffix":""}],"badges":[],"createdAt":"2025-08-30 10:38:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7494692/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7494692/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s11760-025-05066-8","type":"published","date":"2026-01-16T16:28:52+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":100614368,"identity":"3a35ec6c-68be-47b5-85b6-b05fb68d03c6","added_by":"auto","created_at":"2026-01-19 17:19:26","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3544900,"visible":true,"origin":"","legend":"","description":"","filename":"SpringClid.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7494692/v1_covered_9e9d7cb7-73d7-42d9-90b5-2b720656a07e.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Cross-Modal Local Interest Contrast with Dual-Graph Denoising for Multimodal 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":"signal-image-and-video-processing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"sivp","sideBox":"Learn more about [Signal, Image and Video Processing](http://link.springer.com/journal/11760)","snPcode":"11760","submissionUrl":"https://submission.nature.com/new-submission/11760/3","title":"Signal, Image and Video Processing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Multimodal recommendation, Cross-modal alignment, Contrastive learning, Hypergraph learning, Graph denoising.","lastPublishedDoi":"10.21203/rs.3.rs-7494692/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7494692/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eMultimodal recommender systems improve recommendation accuracy by incorporating item multimodal features (e.g., text, images) alongside user-item interactions. However, they face two critical challenges: (1) local user interests in multimodal features are often obscured by irrelevant content (e.g., background clutter in product images), and (2) behavioral data contains pervasive low-credibility interactions (e.g., accidental clicks) that propagate noise through graph-based recommenders. Notably, over-reliance on region-of-interest (ROI) features during graph construction may introduce spurious edges by ignoring global contextual relationships, exacerbating semantic distortion. To address these issues, we propose CLID (Cross-modal Local Interest Denoising), a novel framework integrating \u003cb\u003eC\u003c/b\u003eross-Modal \u003cb\u003eL\u003c/b\u003eocal \u003cb\u003eI\u003c/b\u003enterest Contrast and Dual-Graph \u003cb\u003eD\u003c/b\u003eenoising. First, our local interest contrast mechanism employs text-guided visual attention alignment and a contrastive loss function to enhance discriminative local features\u0026mdash;for example, it learns to focus on \"sleeve design\" in clothing images while suppressing unrelated background features. Crucially, it adaptively weights local features against global representations to prevent ROI-induced bias. Second, the dual-graph denoising architecture combines: (i) a local graph that stabilizes neighbor aggregation via structural consistency to attenuate noisy interactions and (ii) a hypergraph capturing group-wise behavioral patterns, where high-confidence interactions are reinforced through co-occurrence frequency weighting. Experiments demonstrate that CLID significantly improves recommendation performance on three Amazon review datasets: Baby; Clothing, Shoes and Jewelry; as well as Sports and Outdoors. The proposed CLID framework provides a generalizable contrast-and-denoise paradigm for robust multimodal recommendation, effectively bridging fine-grained feature enhancement with noise-resilient graph learning.The code implementation is openly available at the following repository: https://github.com/Qiyx5025/CLID-master.\u003c/p\u003e","manuscriptTitle":"Cross-Modal Local Interest Contrast with Dual-Graph Denoising for Multimodal Recommendation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-26 11:04:43","doi":"10.21203/rs.3.rs-7494692/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-09-11T19:59:46+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-09-02T13:19:26+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-02T13:18:44+00:00","index":"","fulltext":""},{"type":"submitted","content":"Signal, Image and Video Processing","date":"2025-08-30T10:23:45+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"signal-image-and-video-processing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"sivp","sideBox":"Learn more about [Signal, Image and Video Processing](http://link.springer.com/journal/11760)","snPcode":"11760","submissionUrl":"https://submission.nature.com/new-submission/11760/3","title":"Signal, Image and Video Processing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"844fedda-b4e9-45ee-8490-5f5638d7cf05","owner":[],"postedDate":"September 26th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-01-19T16:44:39+00:00","versionOfRecord":{"articleIdentity":"rs-7494692","link":"https://doi.org/10.1007/s11760-025-05066-8","journal":{"identity":"signal-image-and-video-processing","isVorOnly":false,"title":"Signal, Image and Video Processing"},"publishedOn":"2026-01-16 16:28:52","publishedOnDateReadable":"January 16th, 2026"},"versionCreatedAt":"2025-09-26 11:04:43","video":"","vorDoi":"10.1007/s11760-025-05066-8","vorDoiUrl":"https://doi.org/10.1007/s11760-025-05066-8","workflowStages":[]},"version":"v1","identity":"rs-7494692","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7494692","identity":"rs-7494692","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}