Translution: Unifying Transformer and Convolution for Adaptive and Relative Modeling

Translution: Unifying Transformer and Convolution for Adaptive and Relative Modeling | 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 Translution: Unifying Transformer and Convolution for Adaptive and Relative Modeling Hehe Fan, Yi Yang, Fei Wu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6612519/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 When referring to modeling, we consider it to involve two steps: 1) identifying relevant data elements or regions and 2) encoding them effectively. Transformer, leveraging self-attention, can adaptively identify these elements or regions but rely on absolute position encoding for their representation. In contrast, Convolution encodes elements or regions in a relative manner, yet their fixed kernel size limits their ability to adaptively select the relevant regions. We introduce Translution, a new neural network module that unifies the adaptive identification capability of Transformer and the relative encoding advantage of Convolution. However, this integration results in a substantial increase in parameters and memory consumption, exceeding our available computational resources. Therefore, we evaluate Translution on small-scale datasets, i.e., MNIST and CIFAR. Experiments demonstrate that Translution achieves higher accuracy than Transformer. We encourage the community to further evaluate Translution using larger-scale datasets across more diverse scenarios and to develop optimized variants for broader applicability. Artificial Intelligence and Machine Learning Deep neural network Transformer Convolution 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-6612519","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":453319856,"identity":"46bd9533-bcd9-4126-8a14-abaec6745151","order_by":0,"name":"Hehe Fan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA2ElEQVRIiWNgGAWjYBACAwYeBoYKBgY5BoYDQC4bsVrOMDAYk64lsQHMJUaLOXvvwQcHau6kz288Y8DwoewwA//sBvxaLHvOJRscOPYst7HhjAHjjHOHGSTuHCDgsBs5ZtIf2A7nNjOcMWDmbTvMYCCRQEDL/TfmPw78O5zOBtLylygtN3jMGA62HU7gAWlhJErLmRxjiYN9hw1nMBwrONhzLp1H4gYhLcfPGH448O2wvPyMwxsf/CizluOfQUALAkgcAEcmD7HqgYC/gQTFo2AUjIJRMKIAALJrSgjeRkjmAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0001-9572-2345","institution":"College of Computer Science and Technology, Zhejiang University","correspondingAuthor":true,"prefix":"","firstName":"Hehe","middleName":"","lastName":"Fan","suffix":""},{"id":453319857,"identity":"d440c7b7-9bc9-4998-961c-64a13366013f","order_by":1,"name":"Yi Yang","email":"","orcid":"","institution":"College of Computer Science and Technology, Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Yi","middleName":"","lastName":"Yang","suffix":""},{"id":453319858,"identity":"011dc9df-6f72-446d-9a3a-b269a0667098","order_by":2,"name":"Fei Wu","email":"","orcid":"","institution":"College of Computer Science and Technology, Zhejiang University","correspondingAuthor":false,"prefix":"","firstName":"Fei","middleName":"","lastName":"Wu","suffix":""}],"badges":[],"createdAt":"2025-05-07 13:40:50","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":true,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":true},"doi":"10.21203/rs.3.rs-6612519/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6612519/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":82228915,"identity":"7d1dae25-6da0-4ab3-917b-edc1edffea13","added_by":"auto","created_at":"2025-05-08 05:28:53","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":498775,"visible":true,"origin":"","legend":"","description":"","filename":"Translution.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6612519/v1_covered_518f1f57-dd15-4db4-8c2d-1ab5ef9f60b1.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eTranslution: Unifying Transformer and Convolution for Adaptive and Relative Modeling\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Zhejiang University","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"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":"Deep neural network, Transformer, Convolution","lastPublishedDoi":"10.21203/rs.3.rs-6612519/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6612519/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eWhen referring to modeling, we consider it to involve two steps: 1) identifying relevant data elements or regions and 2) encoding them effectively. \u0026nbsp;Transformer, leveraging self-attention, can adaptively identify these elements or regions but rely on absolute position encoding for their representation. In contrast, Convolution encodes elements or regions in a relative manner, yet their fixed kernel size limits their ability to adaptively select the relevant regions. \u0026nbsp;We introduce Translution, a new neural network module that unifies the adaptive identification capability of Transformer and the relative encoding advantage of Convolution. \u0026nbsp;However, this integration results in a substantial increase in parameters and memory consumption, exceeding our available computational resources.\u0026nbsp; Therefore, we \u0026nbsp;evaluate Translution on small-scale \u0026nbsp;datasets, i.e., MNIST and CIFAR. Experiments demonstrate that Translution achieves higher accuracy than Transformer. \u0026nbsp;We encourage the community to further evaluate Translution using larger-scale datasets across \u0026nbsp;more diverse scenarios and to develop optimized variants for broader applicability.\u0026nbsp;\u003c/p\u003e","manuscriptTitle":"Translution: Unifying Transformer and Convolution for Adaptive and Relative Modeling","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-08 05:12:27","doi":"10.21203/rs.3.rs-6612519/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":"c64ffa53-aa09-41e2-bd84-7cae5077f8a2","owner":[],"postedDate":"May 8th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":48209183,"name":"Artificial Intelligence and Machine Learning"}],"tags":[],"updatedAt":"2025-05-08T05:12:27+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-08 05:12:27","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6612519","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6612519","identity":"rs-6612519","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}