Progressive Layer Activation CLIP for Few-Shot and Generalizable Cassava Disease Recognition

Progressive Layer Activation CLIP for Few-Shot and Generalizable Cassava Disease 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 Progressive Layer Activation CLIP for Few-Shot and Generalizable Cassava Disease Recognition Muhammad Shafay, Muhammad Owais, Divya Velayudhan, Taimur Hassan, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9109616/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 Cassava diseases such as Cassava Mosaic Disease (CMD), Cassava Brown Streak Disease (CBSD), and Cassava Bacterial Blight (CBB) pose serious threats to global food security, particularly in resource-limited regions where expert diagnosis is scarce. Although large vision–language models enable automated plant disease recognition, existing fine-tuning approaches struggle under extreme data scarcity. This paper proposes Progressive Layer Activation CLIP (PLA-CLIP), a curriculum-inspired fine-tuning framework for efficient few-shot classification of cassava diseases. PLA-CLIP progressively unfreezes transformer layers during training, stabilizing the optimization process while preserving pretrained vision–language alignment. Using only 43 images per class, PLA-CLIP achieves 78.25% accuracy and a 78.00% F1-weighted score on CD1, outperforming zero-shot CLIP by +15.98% and standard fine-tuning by +3.94%. Cross-dataset evaluations on CD2 and CD3 demonstrate robust generalization across varying conditions. Attention map visualizations confirm that the model focuses on disease-relevant regions, supporting interpretability. With a 2.65 ms inference time and moderate model size, PLA-CLIP offers an effective balance between efficiency and performance for practical plant health monitoring. The implementation and experimental code are publicly available at https://github.com/ mshafay5/PLA-CLIP. Agricultural Engineering Few-shot learning Progressive layer activation CLIP adaptation Attention visualization Full Text Additional Declarations The authors declare no competing interests. Supplementary Files 2FinalPresentationPaperSE103ICCAE2026CameraReadyVersion.pdf Conference Presentation 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-9109616","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":605389659,"identity":"854c868a-69ac-46ee-acd1-e7a073e70f69","order_by":0,"name":"Muhammad Shafay","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3UlEQVRIiWNgGAWjYBACNmbGhsM///2rn8/eRqQWPvbmg48Z2A4wbuw5RqQWOZ5jycYgLQ030oh1mESOmXQBzx1mxpnP0iQYauwY+NsbiNAyQ+IZG7t02jEJhmPJDBJnDhDWIsFjwMzDODu9TQLoQgaGGwnEaElglmC4eRyo5d8BBvn7DwhoAXmf58BhA4YbbMckGNsOMBjcwK+DgQ0YyA9nNqQlGPakJVsk9iXzGJ4h4DD5ZsaGAx8bbBLk2Y8Z3vjwzU5O7vgBAtagAKD5PKSoHwWjYBSMglGAAwAA1FRBoi/4814AAAAASUVORK5CYII=","orcid":"","institution":"Khalifa University","correspondingAuthor":true,"prefix":"","firstName":"Muhammad","middleName":"","lastName":"Shafay","suffix":""},{"id":605389757,"identity":"f1a7159c-92dd-4a8e-912d-49e80d19731a","order_by":1,"name":"Muhammad Owais","email":"","orcid":"","institution":"Khalifa University","correspondingAuthor":false,"prefix":"","firstName":"Muhammad","middleName":"","lastName":"Owais","suffix":""},{"id":605392525,"identity":"723a195c-9d6a-4a26-8b02-20b51590b362","order_by":2,"name":"Divya Velayudhan","email":"","orcid":"","institution":"Khalifa University","correspondingAuthor":false,"prefix":"","firstName":"Divya","middleName":"","lastName":"Velayudhan","suffix":""},{"id":605392526,"identity":"a9d0b9db-d980-4755-92b4-6817c7a4da4f","order_by":3,"name":"Taimur Hassan","email":"","orcid":"","institution":"Abu Dhabi University","correspondingAuthor":false,"prefix":"","firstName":"Taimur","middleName":"","lastName":"Hassan","suffix":""},{"id":605392527,"identity":"e491a045-2cff-4234-9f1e-c7fa731c4deb","order_by":4,"name":"Irfan Hussain","email":"","orcid":"","institution":"Khalifa University","correspondingAuthor":false,"prefix":"","firstName":"Irfan","middleName":"","lastName":"Hussain","suffix":""},{"id":605392528,"identity":"cd5bd9a2-6d11-4792-a67e-90ff7a0b2be4","order_by":5,"name":"Naoufel Werghi","email":"","orcid":"","institution":"Khalifa University","correspondingAuthor":false,"prefix":"","firstName":"Naoufel","middleName":"","lastName":"Werghi","suffix":""}],"badges":[],"createdAt":"2026-03-13 03:09:13","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-9109616/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9109616/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104782932,"identity":"33bba2fd-6484-4584-9056-817651faa163","added_by":"auto","created_at":"2026-03-17 07:57:59","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":843546,"visible":true,"origin":"","legend":"","description":"","filename":"Paper1Shafay.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9109616/v1_covered_28eff3a8-bb35-4cde-8608-59323e748074.pdf"},{"id":104680248,"identity":"8c387c60-be82-4c90-8631-7d653119111c","added_by":"auto","created_at":"2026-03-16 02:14:54","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1040103,"visible":true,"origin":"","legend":"\u003cp\u003eConference Presentation\u003c/p\u003e","description":"","filename":"2FinalPresentationPaperSE103ICCAE2026CameraReadyVersion.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9109616/v1/3c2049667b771e3fcffe204d.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eProgressive Layer Activation CLIP for Few-Shot and Generalizable Cassava Disease Recognition\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Khalifa University of Science and Technology","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":"Few-shot learning, Progressive layer activation, CLIP adaptation, Attention visualization","lastPublishedDoi":"10.21203/rs.3.rs-9109616/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9109616/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eCassava diseases such as Cassava Mosaic Disease (CMD), Cassava Brown Streak Disease (CBSD), and Cassava Bacterial Blight (CBB) pose serious threats to global food security, particularly in resource-limited regions where expert diagnosis is scarce. Although large vision–language models enable automated plant disease recognition, existing fine-tuning approaches struggle under extreme data scarcity. This paper proposes Progressive Layer Activation CLIP (PLA-CLIP), a curriculum-inspired fine-tuning framework for efficient few-shot classification of cassava diseases. PLA-CLIP progressively unfreezes transformer layers during training, stabilizing the optimization process while preserving pretrained vision–language alignment. Using only 43 images per class, PLA-CLIP achieves 78.25% accuracy and a 78.00% F1-weighted score on CD1, outperforming zero-shot CLIP by +15.98% and standard fine-tuning by +3.94%. Cross-dataset evaluations on CD2 and CD3 demonstrate robust generalization across varying conditions. Attention map visualizations confirm that the model focuses on disease-relevant regions, supporting interpretability. With a 2.65 ms inference time and moderate model size, PLA-CLIP offers an effective balance between efficiency and performance for practical plant health monitoring. The implementation and experimental code are publicly available at https://github.com/ mshafay5/PLA-CLIP.\u003c/p\u003e","manuscriptTitle":"Progressive Layer Activation CLIP for Few-Shot and Generalizable Cassava Disease Recognition","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-16 02:14:49","doi":"10.21203/rs.3.rs-9109616/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":"3df705bd-ed6d-4cd5-8489-8fac2f0f1e31","owner":[],"postedDate":"March 16th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":64428566,"name":"Agricultural Engineering"}],"tags":[],"updatedAt":"2026-03-16T02:14:49+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-16 02:14:49","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9109616","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9109616","identity":"rs-9109616","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}