Self-calibrated mutual learning for fine-grained image recognition

Self-calibrated mutual learning for fine-grained image recognition | 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 Self-calibrated mutual learning for fine-grained image recognition Jung-Ha Hwang, Doo-Hyun Choi This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8942212/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 13 You are reading this latest preprint version Abstract Knowledge distillation has demonstrated its effectiveness in fine-grained image classification. While recent knowledge distillation methods typically improve classification accuracy by increasing the number of models or exploiting intermediate representations, they often require extensive training time or architectural modifications and insufficiently consider the geometry of output distributions in fine-grained regimes. To address this issue, this paper proposes a method for enhancing fine-grained classification accuracy based on the integration of mutual learning and self-distillation. This method achieves improved accuracy by transforming the geometry of output distributions through the fusion of cross-model consistency and self-calibration. Cross-model consistency enhances generalizability by sharing peer knowledge, and self-calibration strengthens intra-class similarities by incurring overconfidence suppression. The proposed method is validated on three benchmark datasets across multiple backbone architectures, compared with existing methods. Experimental results demonstrate that the proposed method improves accuracy over existing methods and exhibits a complementary effect beyond a linear combination of mutual learning and self-distillation. Fine-grained image classification Knowledge distillation Mutual learning Self-distillation Hybrid mutual learning Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 12 Mar, 2026 Reviews received at journal 11 Mar, 2026 Reviews received at journal 10 Mar, 2026 Reviewers agreed at journal 06 Mar, 2026 Reviewers agreed at journal 05 Mar, 2026 Reviewers agreed at journal 05 Mar, 2026 Reviews received at journal 05 Mar, 2026 Reviewers agreed at journal 04 Mar, 2026 Reviewers agreed at journal 04 Mar, 2026 Reviewers invited by journal 04 Mar, 2026 Editor assigned by journal 28 Feb, 2026 Submission checks completed at journal 28 Feb, 2026 First submitted to journal 22 Feb, 2026 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-8942212","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":601805354,"identity":"ea02f796-3bed-4ad1-8ec3-6f16c04ccdb7","order_by":0,"name":"Jung-Ha Hwang","email":"","orcid":"","institution":"Kyungpook National University","correspondingAuthor":false,"prefix":"","firstName":"Jung-Ha","middleName":"","lastName":"Hwang","suffix":""},{"id":601805355,"identity":"3c8bfbd8-3c23-4cdc-b491-03f3b340b5b7","order_by":1,"name":"Doo-Hyun Choi","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAuElEQVRIiWNgGAWjYDCCGwyMD4CUAZSbQJQWZgOStbBJkKaF73bzs4oPFQzG5hIJjB9+MKTlE9QieeeY2c0ZZxjMLGckMEv2MORYNhDSYnAjh+02bxuDjcGNBAZpBoYKA0I6wFqK//4Da2H+TbQWZsYGBjOgFjagLTmEtUjeSDOW7DkmYWxw5mGbZY9BGmEtfDeSH374UWNjuOF48uEbPyqSCWuBAlDUAJ3HQLSGUTAKRsEoGAV4AQC+zDfjIQVo9wAAAABJRU5ErkJggg==","orcid":"","institution":"Kyungpook National University","correspondingAuthor":true,"prefix":"","firstName":"Doo-Hyun","middleName":"","lastName":"Choi","suffix":""}],"badges":[],"createdAt":"2026-02-23 02:09:00","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8942212/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8942212/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104404848,"identity":"11e6980a-c95c-4649-ab4e-c27e4e30ca8f","added_by":"auto","created_at":"2026-03-11 12:21:12","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":610828,"visible":true,"origin":"","legend":"","description":"","filename":"ManuscriptNPL.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8942212/v1_covered_9ddb85ca-3726-4cdb-bfde-0f9e0dcede15.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Self-calibrated mutual learning for fine-grained image recognition","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":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"neural-processing-letters","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"nepl","sideBox":"Learn more about [Neural Processing Letters](http://link.springer.com/journal/11063)","snPcode":"11063","submissionUrl":"https://submission.nature.com/new-submission/11063/3","title":"Neural Processing Letters","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Fine-grained image classification, Knowledge distillation, Mutual learning, Self-distillation, Hybrid mutual learning","lastPublishedDoi":"10.21203/rs.3.rs-8942212/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8942212/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eKnowledge distillation has demonstrated its effectiveness in fine-grained image classification. While recent knowledge distillation methods typically improve classification accuracy by increasing the number of models or exploiting intermediate representations, they often require extensive training time or architectural modifications and insufficiently consider the geometry of output distributions in fine-grained regimes. To address this issue, this paper proposes a method for enhancing fine-grained classification accuracy based on the integration of mutual learning and self-distillation. This method achieves improved accuracy by transforming the geometry of output distributions through the fusion of cross-model consistency and self-calibration. Cross-model consistency enhances generalizability by sharing peer knowledge, and self-calibration strengthens intra-class similarities by incurring overconfidence suppression. The proposed method is validated on three benchmark datasets across multiple backbone architectures, compared with existing methods. Experimental results demonstrate that the proposed method improves accuracy over existing methods and exhibits a complementary effect beyond a linear combination of mutual learning and self-distillation.\u003c/p\u003e","manuscriptTitle":"Self-calibrated mutual learning for fine-grained image recognition","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-09 05:14:15","doi":"10.21203/rs.3.rs-8942212/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-03-12T21:23:27+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-11T11:57:52+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-10T08:55:14+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"67970012075858660399849042753241420612","date":"2026-03-06T10:21:19+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"311791038615805790894090069991144119863","date":"2026-03-06T03:57:35+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"193113331954931077495379566634738142794","date":"2026-03-06T01:17:26+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-05T16:46:00+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"74775840899014295155214447263601907249","date":"2026-03-04T16:35:24+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"55629947535131431781237565425459221056","date":"2026-03-04T06:11:22+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-04T06:02:30+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-28T12:52:20+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-28T12:49:00+00:00","index":"","fulltext":""},{"type":"submitted","content":"Neural Processing Letters","date":"2026-02-23T01:58:12+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"neural-processing-letters","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"nepl","sideBox":"Learn more about [Neural Processing Letters](http://link.springer.com/journal/11063)","snPcode":"11063","submissionUrl":"https://submission.nature.com/new-submission/11063/3","title":"Neural Processing Letters","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"d083ca86-972b-4e0a-9394-045d80929878","owner":[],"postedDate":"March 9th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-14T21:38:07+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-09 05:14:15","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8942212","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8942212","identity":"rs-8942212","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}