TKANMiXer: Multi-scale hybrid KAN for long-term forecasting

TKANMiXer: Multi-scale hybrid KAN for long-term forecasting | 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 TKANMiXer: Multi-scale hybrid KAN for long-term forecasting Yuting Kong, Huijia Zhao, Hua Zheng, Xiaoping Lu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7325494/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 17 Feb, 2026 Read the published version in Soft Computing → Version 1 posted 5 You are reading this latest preprint version Abstract Time series forecasting is extensively applied across diverse real-world domains, such as finance, power systems, and meteorology, yet its accuracy is often constrained by the inherent complexity of temporal data. While exogenous variables can enhance predictive performance, their inclusion introduces additional challenges by increasing the intricacy of the time series structure. Moreover, traditional Multi-Layer Perceptrons (MLPs) frequently fail to effectively capture hierarchical feature representations across multiple scales, further complicating time series modeling. To address these limitations, we propose TKANMiXer, a novel Kolmogorov-Arnold Network(KAN)-based framework designed for variable fusion and multi-scale decomposition. Our framework comprises two key components: (1) incorporating a seasonal-trend decomposition module on the basis of variable fusion, (2) the multi-layer KAN architecture hierarchically captures. Specifically, TKANMiXer addresses the complexity arising from exogenous variables by disentangling temporal patterns, multi-layer KAN architecture hierarchically captures multi-scale temporal dependencies, significantly improving forecasting accuracy. Extensive experiments demonstrate that TKANMiXer achieves state-of-the-art performance in long-term forecasting tasks across multiple benchmark datasets. Time series prediction Deep learning Attention mechanism Seasonal trend decomposition Kolmogorov-Arnold Network Full Text Cite Share Download PDF Status: Published Journal Publication published 17 Feb, 2026 Read the published version in Soft Computing → Version 1 posted Editorial decision: Major Revision 27 Oct, 2025 Reviewers agreed at journal 30 Aug, 2025 Reviewers invited by journal 13 Aug, 2025 Editor invited by journal 13 Aug, 2025 First submitted to journal 08 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-7325494","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":500168583,"identity":"d49a6875-d29e-4e16-8973-2e469458de50","order_by":0,"name":"Yuting Kong","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA+UlEQVRIiWNgGAWjYBACCSjNw8DAfABEgoABMVoMgIrZEkBaJIjWArLIgDgtku29h1/ztv2RMedf8/nF27a6Ogb25m0SDDV3cGqR5jmXZs3bZsBjOePtNsu5bYclGHiOlUkwHHuGU4ucRI6ZMUiLwY2z24CMAxIMQBEJxobDxGg58wzIqJNgkH+DX4u0RI7xY7CW8z3MQAYz0BYe/Foke86YMc45Zwy0hQ3EOCzZxpNWbJFwDLcWieM9xh/elMnZG5w//BjIqOPnZz+88caHGtxagIBNChzpEgls4EhiAxEJ+DQAE8rHHyCK/wDzB/wKR8EoGAWjYKQCACU0TzcJA2ePAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0009-0000-8568-4808","institution":"Guangdong University of Science and Technology","correspondingAuthor":true,"prefix":"","firstName":"Yuting","middleName":"","lastName":"Kong","suffix":""},{"id":500168584,"identity":"e92af0f8-0fa4-472f-8969-27350800cc78","order_by":1,"name":"Huijia Zhao","email":"","orcid":"","institution":"Macau University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Huijia","middleName":"","lastName":"Zhao","suffix":""},{"id":500168585,"identity":"2bf836d5-5d75-4a2a-b580-85fc11aabdee","order_by":2,"name":"Hua Zheng","email":"","orcid":"","institution":"Shaoguan University","correspondingAuthor":false,"prefix":"","firstName":"Hua","middleName":"","lastName":"Zheng","suffix":""},{"id":500168586,"identity":"e489b55a-715f-4d5f-8fee-cfdecc22e149","order_by":3,"name":"Xiaoping Lu","email":"","orcid":"","institution":"Macau University of Science and Technology","correspondingAuthor":false,"prefix":"","firstName":"Xiaoping","middleName":"","lastName":"Lu","suffix":""}],"badges":[],"createdAt":"2025-08-08 09:04:36","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7325494/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7325494/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s00500-025-11061-5","type":"published","date":"2026-02-17T15:59:11+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":103252361,"identity":"c433a2cf-7d58-42f7-86f6-d108ebc86260","added_by":"auto","created_at":"2026-02-23 16:14:31","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1265710,"visible":true,"origin":"","legend":"","description":"","filename":"TKANMiXerMultiscalehybridKANforlongtermforecasting.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7325494/v1_covered_64d542da-34e4-4285-837b-efa3cf58774c.pdf"}],"financialInterests":"","formattedTitle":"TKANMiXer: Multi-scale hybrid KAN for long-term forecasting","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":true,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"soft-computing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"soco","sideBox":"Learn more about [Soft Computing](https://www.springer.com/journal/500)","snPcode":"500","submissionUrl":"https://submission.nature.com/new-submission/500/3","title":"Soft Computing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Time series prediction, Deep learning, Attention mechanism, Seasonal trend decomposition, Kolmogorov-Arnold Network","lastPublishedDoi":"10.21203/rs.3.rs-7325494/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7325494/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Time series forecasting is extensively applied across diverse real-world domains, such as finance, power systems, and meteorology, yet its accuracy is often constrained by the inherent complexity of temporal data. While exogenous variables can enhance predictive performance, their inclusion introduces additional challenges by increasing the intricacy of the time series structure. Moreover, traditional Multi-Layer Perceptrons (MLPs) frequently fail to effectively capture hierarchical feature representations across multiple scales, further complicating time series modeling. To address these limitations, we propose TKANMiXer, a novel Kolmogorov-Arnold Network(KAN)-based framework designed for variable fusion and multi-scale decomposition. Our framework comprises two key components: (1) incorporating a seasonal-trend decomposition module on the basis of variable fusion, (2) the multi-layer KAN architecture hierarchically captures. Specifically, TKANMiXer addresses the complexity arising from exogenous variables by disentangling temporal patterns, multi-layer KAN architecture hierarchically captures multi-scale temporal dependencies, significantly improving forecasting accuracy. Extensive experiments demonstrate that TKANMiXer achieves state-of-the-art performance in long-term forecasting tasks across multiple benchmark datasets.","manuscriptTitle":"TKANMiXer: Multi-scale hybrid KAN for long-term forecasting","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-21 15:41:43","doi":"10.21203/rs.3.rs-7325494/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Major Revision","date":"2025-10-28T01:49:52+00:00","index":"","fulltext":""},{"type":"reviewerAgreed","content":"","date":"2025-08-30T07:04:42+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-13T21:12:18+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"Soft Computing","date":"2025-08-13T19:07:27+00:00","index":"","fulltext":""},{"type":"submitted","content":"Soft Computing","date":"2025-08-08T05:04:28+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"soft-computing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"soco","sideBox":"Learn more about [Soft Computing](https://www.springer.com/journal/500)","snPcode":"500","submissionUrl":"https://submission.nature.com/new-submission/500/3","title":"Soft Computing","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"afd76118-017e-40b1-98c6-f0350b582f55","owner":[],"postedDate":"August 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2026-02-23T16:12:37+00:00","versionOfRecord":{"articleIdentity":"rs-7325494","link":"https://doi.org/10.1007/s00500-025-11061-5","journal":{"identity":"soft-computing","isVorOnly":false,"title":"Soft Computing"},"publishedOn":"2026-02-17 15:59:11","publishedOnDateReadable":"February 17th, 2026"},"versionCreatedAt":"2025-08-21 15:41:43","video":"","vorDoi":"10.1007/s00500-025-11061-5","vorDoiUrl":"https://doi.org/10.1007/s00500-025-11061-5","workflowStages":[]},"version":"v1","identity":"rs-7325494","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7325494","identity":"rs-7325494","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}