Collaborative Mixture of Experts for Enhanced Code-Switching Visual Recognition Through Recursive Token Propagation

Collaborative Mixture of Experts for Enhanced Code-Switching Visual Recognition Through Recursive Token Propagation | 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 Collaborative Mixture of Experts for Enhanced Code-Switching Visual Recognition Through Recursive Token Propagation Liam Nisapo, Daniel Garcia, Olivia Williams, Taylor Prence This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5038514/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 Multilingual environments often require models to accurately recognize and process both linguistic and visual inputs, particularly in cases where code-switching occurs frequently. Traditional models designed for language and visual recognition struggle to handle such complexities due to their inability to dynamically allocate resources to specific tasks. The introduction of a Collaborative Mixture of Experts (MoE) model within Mistral 8x7b addresses this limitation through the implementation of expert modules that specialize in different aspects of code-switching and visual recognition. A dynamic gating mechanism selects the most appropriate expert based on the input, leading to substantial improvements in precision, recall, and cross-modal accuracy. Experimental results demonstrate that the MoE architecture not only enhances linguistic boundary detection in multilingual contexts but also significantly improves the alignment between language and visual data, ensuring more robust and adaptive performance in real-time applications. The findings indicate that the proposed model outperforms baseline models in both accuracy and computational efficiency, suggesting its potential for further advancements in handling complex multimodal tasks. Artificial Intelligence and Machine Learning Code-switching visual recognition Mixture of Experts multimodal processing multilingual 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-5038514","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":350039667,"identity":"bce6900b-0947-4c42-9b16-414bd957e2c5","order_by":0,"name":"Liam Nisapo","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA6UlEQVRIiWNgGAWjYDACZjBpY8fG3nyYJC1pyfw8x5JBLAli7TrMOHNGjjFxWszZ2R9+LqhIYzY4c+azwccdNnX8/QcYP3zMwa3FspnHWHrGGRs+g+O9mxNnnkmTkLiRwCw5cxtuLQaHeRikedtAtpzdfJi37bAEww0GNmZevFrYH/8GqmTccCPnMVDLfwn58wcIaWEwkwZpAXqfOZm37YCEwYEEQlp4zKxnnAEHsrHhzLZkyY03Epvx++X88ce3CyrAUflY4mObHb/c+cMHP3zEowUEmNH4jA341WPRMgpGwSgYBaMAFQAAxQdRkqfw9t0AAAAASUVORK5CYII=","orcid":"https://orcid.org/0009-0009-8584-5742","institution":"","correspondingAuthor":true,"prefix":"","firstName":"Liam","middleName":"","lastName":"Nisapo","suffix":""},{"id":350039669,"identity":"4c897168-78ee-400f-9dc4-0ea20aac022d","order_by":1,"name":"Daniel Garcia","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Daniel","middleName":"","lastName":"Garcia","suffix":""},{"id":350039670,"identity":"2da0682b-a457-45f9-8ec3-20b7bad47974","order_by":2,"name":"Olivia Williams","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Olivia","middleName":"","lastName":"Williams","suffix":""},{"id":350039671,"identity":"3c164f26-a45a-4077-8f7c-1d15f7ae0c8a","order_by":3,"name":"Taylor Prence","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Taylor","middleName":"","lastName":"Prence","suffix":""}],"badges":[],"createdAt":"2024-09-05 13:28:11","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-5038514/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5038514/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":64062369,"identity":"5a7d7c97-340b-4fad-a351-127489fce715","added_by":"auto","created_at":"2024-09-06 02:18:28","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":201752,"visible":true,"origin":"","legend":"","description":"","filename":"moe.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5038514/v1_covered_98dba5d8-e1da-400f-a427-368286ea684c.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eCollaborative Mixture of Experts for Enhanced Code-Switching Visual Recognition Through Recursive Token Propagation\u003c/p\u003e","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":"Code-switching, visual recognition, Mixture of Experts, multimodal processing, multilingual","lastPublishedDoi":"10.21203/rs.3.rs-5038514/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5038514/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eMultilingual environments often require models to accurately recognize and process both linguistic and visual inputs, particularly in cases where code-switching occurs frequently. Traditional models designed for language and visual recognition struggle to handle such complexities due to their inability to dynamically allocate resources to specific tasks. The introduction of a Collaborative Mixture of Experts (MoE) model within Mistral 8x7b addresses this limitation through the implementation of expert modules that specialize in different aspects of code-switching and visual recognition. A dynamic gating mechanism selects the most appropriate expert based on the input, leading to substantial improvements in precision, recall, and cross-modal accuracy. Experimental results demonstrate that the MoE architecture not only enhances linguistic boundary detection in multilingual contexts but also significantly improves the alignment between language and visual data, ensuring more robust and adaptive performance in real-time applications. The findings indicate that the proposed model outperforms baseline models in both accuracy and computational efficiency, suggesting its potential for further advancements in handling complex multimodal tasks.\u003c/p\u003e","manuscriptTitle":"Collaborative Mixture of Experts for Enhanced Code-Switching Visual Recognition Through Recursive Token Propagation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-09-06 02:10:21","doi":"10.21203/rs.3.rs-5038514/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":"04ab45a8-e993-47c7-beb2-569a67321845","owner":[],"postedDate":"September 6th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":37156477,"name":"Artificial Intelligence and Machine Learning"}],"tags":[],"updatedAt":"2024-09-06T02:10:22+00:00","versionOfRecord":[],"versionCreatedAt":"2024-09-06 02:10:21","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-5038514","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5038514","identity":"rs-5038514","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}