M-S4DLite: Orthogonal Decoupling for Lightweight Time-Series Classification via Diagonal State-Space Models

M-S4DLite: Orthogonal Decoupling for Lightweight Time-Series Classification via Diagonal State-Space Models | 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 M-S4DLite: Orthogonal Decoupling for Lightweight Time-Series Classification via Diagonal State-Space Models jian chen, yunfei han, yi wang, shiwei guo, yupeng ma This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8919991/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 Emerging state space models have revolutionized time series forecasting by efficiently capturing long range dependencies, yet their application to classification tasks reveals a fundamental semantic misalignment. Rooted in continuous control theory, standard models suffer from spectral oversmoothing where continuous inductive bias tends to erode the sharp and discrete discriminative motifs critical for classification. To resolve this theoretical conflict, we propose M-S4DLite, a lightweight framework built on the principle of Orthogonal Decoupling. By structurally separating local feature extraction from global evolution, we employ a discrete patching module to preserve high frequency transients and synergize it with a diagonalized S4D backbone for global reasoning. This architecture effectively recalibrates the inductive bias to reconcile discrete local precision with continuous global context while maintaining quasi linear complexity. Extensive experiments demonstrate that M-S4DLite establishes a new Pareto frontier in the balance between efficiency and accuracy, consistently achieving superior performance on diverse benchmarks and exhibiting robustness against resolution degradation compared to pure continuous or convolutional baselines. Physical sciences/Engineering Physical sciences/Mathematics and computing Time-Series Classification State-Space Models Orthogonal Decoupling Spectral Oversmoothing Inductive Bias Calibration Lightweight Deep Learning 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-8919991","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":599599534,"identity":"bd0cb85a-126e-4661-9b5b-ff4fdc4f536f","order_by":0,"name":"jian chen","email":"data:image/png;base64,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","orcid":"","institution":"","correspondingAuthor":true,"prefix":"","firstName":"jian","middleName":"","lastName":"chen","suffix":""},{"id":599599535,"identity":"02e8db11-bf28-4a06-98c7-41ac2d40bf42","order_by":1,"name":"yunfei han","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"yunfei","middleName":"","lastName":"han","suffix":""},{"id":599599537,"identity":"4e0336bb-a7ab-401f-ade0-8a682019c7ec","order_by":2,"name":"yi wang","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"yi","middleName":"","lastName":"wang","suffix":""},{"id":599599539,"identity":"cba57671-cc43-41f8-9188-bfc3d7cff09b","order_by":3,"name":"shiwei guo","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"shiwei","middleName":"","lastName":"guo","suffix":""},{"id":599599541,"identity":"1ba2e6ea-5874-470c-a943-8717b622fb76","order_by":4,"name":"yupeng ma","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"yupeng","middleName":"","lastName":"ma","suffix":""}],"badges":[],"createdAt":"2026-02-19 18:08:27","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8919991/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8919991/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":109176733,"identity":"dcda9acb-d503-4bb6-93c2-b4bd259af3c2","added_by":"auto","created_at":"2026-05-13 09:33:16","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":811396,"visible":true,"origin":"","legend":"","description":"","filename":"Ms4Dlite.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8919991/v1_covered_76fe805f-b28f-4786-b50f-359749858cba.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"M-S4DLite: Orthogonal Decoupling for Lightweight Time-Series Classification via Diagonal State-Space Models","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":"Time-Series Classification, State-Space Models, Orthogonal Decoupling, Spectral Oversmoothing, Inductive Bias Calibration, Lightweight Deep Learning","lastPublishedDoi":"10.21203/rs.3.rs-8919991/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8919991/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eEmerging state space models have revolutionized time series forecasting by efficiently capturing long range dependencies, yet their application to classification tasks reveals a fundamental semantic misalignment. Rooted in continuous control theory, standard models suffer from spectral oversmoothing where continuous inductive bias tends to erode the sharp and discrete discriminative motifs critical for classification. To resolve this theoretical conflict, we propose M-S4DLite, a lightweight framework built on the principle of Orthogonal Decoupling. By structurally separating local feature extraction from global evolution, we employ a discrete patching module to preserve high frequency transients and synergize it with a diagonalized S4D backbone for global reasoning. This architecture effectively recalibrates the inductive bias to reconcile discrete local precision with continuous global context while maintaining quasi linear complexity. Extensive experiments demonstrate that M-S4DLite establishes a new Pareto frontier in the balance between efficiency and accuracy, consistently achieving superior performance on diverse benchmarks and exhibiting robustness against resolution degradation compared to pure continuous or convolutional baselines.\u003c/p\u003e","manuscriptTitle":"M-S4DLite: Orthogonal Decoupling for Lightweight Time-Series Classification via Diagonal State-Space Models","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-03 17:59:26","doi":"10.21203/rs.3.rs-8919991/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":"7a1dafa3-ceb1-45f8-b9ce-0def20babb34","owner":[],"postedDate":"March 3rd, 2026","published":true,"recentEditorialEvents":[{"type":"decision","content":"Withdrawn","date":"2026-05-13T09:20:13+00:00","index":"","fulltext":""},{"type":"decision","content":"Revision requested","date":"2026-05-06T09:47:50+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-06T01:55:29+00:00","index":265,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":63803251,"name":"Physical sciences/Engineering"},{"id":63803252,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-05-13T09:32:07+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-03 17:59:26","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8919991","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8919991","identity":"rs-8919991","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}