Enhancing ConvNeXt for efficient small-size image classification

Enhancing ConvNeXt for efficient small-size image classification | 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 Enhancing ConvNeXt for efficient small-size image classification Jianwei Feng, Jinguo Mo, Hengliang Tan, Shuo Yang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7965188/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 13 You are reading this latest preprint version Abstract Vision Transformers (ViTs) have achieved remarkable success in computer vision, particularly with the advent of the Swin Transformer (Swin-T). Recently, ConvNeXt was proposed to revisit the architecture of convolutional neural networks by incorporating techniques from Swin-T, which achieves competitive performance. However, ConvNeXt exhibited relatively lower accuracy and efficiency on small-size datasets with low-resolution images. To address this issue, based on the structure of ConvNeXt, we propose to employ small kernel sizes to better capture features from low-resolution images. A symmetrical inverted bottleneck structure inspired by MobileNet is used to refine the traditional residual block design. We combine the batch normalization and layer normalization to strengthen the relationships between different samples. We use a twice-patch embedding methodology to dynamically generate adaptive patch sizes according to different input image sizes. Subsequently, we integrate the global response normalization to amplify the contrast and specificity of multiple channels for learning diverse features. Finally, we introduce a novel spatial coordinate attention mechanism that effectively captures global feature information. The proposed method demonstrates superior performance on small-size datasets of CIFAR-10, CIFAR-100, Tiny-ImageNet and Fashion-MNIST with fewer parameters, which confirms the effectiveness and efficiency of our approach. Convolutional neural network Symmetrical inverted bottleneck Spatial coordinate attention Global response normalization Image classification Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 06 Mar, 2026 Reviews received at journal 28 Feb, 2026 Reviewers agreed at journal 07 Feb, 2026 Reviews received at journal 13 Dec, 2025 Reviews received at journal 10 Dec, 2025 Reviewers agreed at journal 10 Dec, 2025 Reviewers agreed at journal 08 Dec, 2025 Reviews received at journal 28 Nov, 2025 Reviewers agreed at journal 11 Nov, 2025 Reviewers invited by journal 06 Nov, 2025 Editor assigned by journal 01 Nov, 2025 Submission checks completed at journal 29 Oct, 2025 First submitted to journal 27 Oct, 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-7965188","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":545737812,"identity":"afe353a0-bd61-434c-841c-7c9c15d0712d","order_by":0,"name":"Jianwei Feng","email":"","orcid":"","institution":"Guangzhou University","correspondingAuthor":false,"prefix":"","firstName":"Jianwei","middleName":"","lastName":"Feng","suffix":""},{"id":545737813,"identity":"0b62130f-a8e8-43d8-9266-21fc628b0c2b","order_by":1,"name":"Jinguo Mo","email":"","orcid":"","institution":"Guangzhou University","correspondingAuthor":false,"prefix":"","firstName":"Jinguo","middleName":"","lastName":"Mo","suffix":""},{"id":545737814,"identity":"2cc25b99-61cc-4d11-8b03-eaa08b8f4af5","order_by":2,"name":"Hengliang Tan","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCUlEQVRIiWNgGAWjYHACNhAhAyY/MDAkgGgJfOp5oFp4QCTjDJK0gAhmHmK02LOfPfbg445aHj72s4df2/yqyzM4wHzwNg+DXR5OW3jy0g1nnjnOw8aTl2ad28dWbHCALdmahyG5GLfDcsykeduOAf2SY2ac28OTuOEAj5k0D8OBxAZcWvjfmEn/BWkBMowteySAWvi/4dciAbSFsa2Gh00ix/gxww8DkC1s+LXceGMm2dsGVCbxxoyxtyEhceZhNmPLOQbJOLWw9+eYSfxsq5OT788x/vDjT11i3/HmhzfeVNjh1AIFh0EEmwRjG5BiBrEN8KsHgjoQwfyB4Q9BlaNgFIyCUTACAQCO4lEQ05uc8AAAAABJRU5ErkJggg==","orcid":"","institution":"Guangzhou University","correspondingAuthor":true,"prefix":"","firstName":"Hengliang","middleName":"","lastName":"Tan","suffix":""},{"id":545737815,"identity":"757c8524-198e-461c-b235-a77edc621463","order_by":3,"name":"Shuo Yang","email":"","orcid":"","institution":"Guangzhou University","correspondingAuthor":false,"prefix":"","firstName":"Shuo","middleName":"","lastName":"Yang","suffix":""}],"badges":[],"createdAt":"2025-10-28 12:06:38","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7965188/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7965188/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96051305,"identity":"850b00e9-ec76-4bcd-a7c4-edbba1a848ed","added_by":"auto","created_at":"2025-11-17 06:40:25","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5749,"visible":true,"origin":"","legend":"","description":"","filename":"31a8175758ff49c1815f70f1c6cd8a27.json","url":"https://assets-eu.researchsquare.com/files/rs-7965188/v1/06c15b8741779eac6206380b.json"},{"id":96246156,"identity":"2c39fd56-3b9b-4ff8-ab1c-9391620e67fe","added_by":"auto","created_at":"2025-11-19 07:24:52","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3933728,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7965188/v1_covered_dab98368-3e1a-4901-a902-95ce5374f98b.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Enhancing ConvNeXt for efficient small-size image classification","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"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":"international-journal-of-machine-learning-and-cybernetics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jmlc","sideBox":"Learn more about [International Journal of Machine Learning and Cybernetics](http://actavetscand.biomedcentral.com/)","snPcode":"13042","submissionUrl":"https://submission.nature.com/new-submission/13042/3","title":"International Journal of Machine Learning and Cybernetics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Convolutional neural network, Symmetrical inverted bottleneck, Spatial coordinate attention, Global response normalization, Image classification","lastPublishedDoi":"10.21203/rs.3.rs-7965188/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7965188/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Vision Transformers (ViTs) have achieved remarkable success in computer vision, particularly with the advent of the Swin Transformer (Swin-T). Recently, ConvNeXt was proposed to revisit the architecture of convolutional neural networks by incorporating techniques from Swin-T, which achieves competitive performance. However, ConvNeXt exhibited relatively lower accuracy and efficiency on small-size datasets with low-resolution images. To address this issue, based on the structure of ConvNeXt, we propose to employ small kernel sizes to better capture features from low-resolution images. A symmetrical inverted bottleneck structure inspired by MobileNet is used to refine the traditional residual block design. We combine the batch normalization and layer normalization to strengthen the relationships between different samples. We use a twice-patch embedding methodology to dynamically generate adaptive patch sizes according to different input image sizes. Subsequently, we integrate the global response normalization to amplify the contrast and specificity of multiple channels for learning diverse features. Finally, we introduce a novel spatial coordinate attention mechanism that effectively captures global feature information. The proposed method demonstrates superior performance on small-size datasets of CIFAR-10, CIFAR-100, Tiny-ImageNet and Fashion-MNIST with fewer parameters, which confirms the effectiveness and efficiency of our approach.","manuscriptTitle":"Enhancing ConvNeXt for efficient small-size image classification","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-17 06:40:21","doi":"10.21203/rs.3.rs-7965188/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-03-06T06:20:48+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-01T03:03:09+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"183705847188266616045700928144286745426","date":"2026-02-08T04:18:48+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-14T00:30:49+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-10T12:11:28+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"340267668201881114922986082718081609641","date":"2025-12-10T11:59:55+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"201982934000781856025497770525997995299","date":"2025-12-08T11:31:12+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-28T16:41:32+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"245155897031904953541480417988032589920","date":"2025-11-11T09:41:56+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-11-06T09:30:59+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-11-01T08:06:51+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-29T13:06:10+00:00","index":"","fulltext":""},{"type":"submitted","content":"International Journal of Machine Learning and Cybernetics","date":"2025-10-28T03:33:50+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"international-journal-of-machine-learning-and-cybernetics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"jmlc","sideBox":"Learn more about [International Journal of Machine Learning and Cybernetics](http://actavetscand.biomedcentral.com/)","snPcode":"13042","submissionUrl":"https://submission.nature.com/new-submission/13042/3","title":"International Journal of Machine Learning and Cybernetics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"98b982dd-0abc-4075-89e3-bdf5b9af57d2","owner":[],"postedDate":"November 17th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[],"tags":[],"updatedAt":"2026-03-06T06:24:55+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-17 06:40:21","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7965188","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7965188","identity":"rs-7965188","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}