Degradation-Guided Prompting and Dynamic Task Curriculum for All-in-One Image Restoration

Degradation-Guided Prompting and Dynamic Task Curriculum for All-in-One Image Restoration | 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 Degradation-Guided Prompting and Dynamic Task Curriculum for All-in-One Image Restoration Dongyang Zhang, Zhijie Zhang, Shenqi Wang, Lya Fu, Junmin Liu This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9298008/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 10 You are reading this latest preprint version Abstract All-in-one image restoration has emerged as a pivotal paradigm in low-level vision. Unlike conventional single-task approaches, this paradigm aims to learn a unified model capable of handling multiple degradation types within a single framework. However, most existing methods predominantly suffer from two critical limitations: i) cross-degradation knowledge contamination during joint optimization, and ii) catastrophic forgetting in sequential learning scenarios. To address these limitations, we introduce a Degradation-Guided Prompt Learning Network (DPLNet), a unified architecture that employs a plug-and-play prompt block to explicitly mine degradation-specific priors from the input, thereby injecting task-adaptive guidance into the restoration process. Complementing the architecture, we introduce the Dynamic Task Curriculum (DTC), a principled learning strategy that sequentially schedules multiple tasks within each training epoch. DTC quantifies per-instance hardness through an online uncertainty criterion and progressively enriches each task optimization by retaining the hard examples from previous tasks, yielding continual robustness gains. Extensive experiments demonstrate that DPLNet achieves state-of-the-art performance in both single-task and multi-task settings. And our proposed training strategy yields consistent improvements across existing architectures. All-in-one image restoration multiple degradation types degradation-guided prompt learning dynamic task curriculum Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 18 May, 2026 Reviews received at journal 28 Apr, 2026 Reviewers agreed at journal 20 Apr, 2026 Reviewers agreed at journal 20 Apr, 2026 Reviewers agreed at journal 20 Apr, 2026 Reviewers agreed at journal 20 Apr, 2026 Reviewers invited by journal 20 Apr, 2026 Editor assigned by journal 06 Apr, 2026 Submission checks completed at journal 02 Apr, 2026 First submitted to journal 02 Apr, 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-9298008","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":627436189,"identity":"edc1d98f-e08c-4d5d-81af-c23baa82c705","order_by":0,"name":"Dongyang Zhang","email":"","orcid":"","institution":"Xi'an Jiaotong University","correspondingAuthor":false,"prefix":"","firstName":"Dongyang","middleName":"","lastName":"Zhang","suffix":""},{"id":627436191,"identity":"c222622a-acc6-4be5-9fa6-9e133fd236b2","order_by":1,"name":"Zhijie Zhang","email":"","orcid":"","institution":"Xi'an Jiaotong University","correspondingAuthor":false,"prefix":"","firstName":"Zhijie","middleName":"","lastName":"Zhang","suffix":""},{"id":627436192,"identity":"f59b153e-1488-4ad2-a351-d75df462ff1e","order_by":2,"name":"Shenqi Wang","email":"","orcid":"","institution":"Xi'an Jiaotong University","correspondingAuthor":false,"prefix":"","firstName":"Shenqi","middleName":"","lastName":"Wang","suffix":""},{"id":627436194,"identity":"79cbc363-f9eb-44f3-bbf4-855af89895da","order_by":3,"name":"Lya Fu","email":"","orcid":"","institution":"Xi'an Jiaotong University","correspondingAuthor":false,"prefix":"","firstName":"Lya","middleName":"","lastName":"Fu","suffix":""},{"id":627436196,"identity":"44b6ae0c-e6c5-4aec-873e-3711388f169b","order_by":4,"name":"Junmin Liu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA/0lEQVRIiWNgGAWjYBACxgYgkcBgwcPADmQlFEBFGNgIapHgYeA5AGQYAEXaCGiBAgkgSgDSBjDVeLQwz8g9uuHhDgkZg5uvEz88MGBIbJ7fY8DwoewwA//sBuwOm5GXdiPxjASPwe3czRJAhxkztvEYMM44d5hB4s4BHFpyzG4ktoG1bABpkQNpYeZtO8xgAHYqPi03z27+AdTCA9bylygtN3i3IWxhxKel5w1Ei+SZ3G0WCQYSQL+kFRzsOZfOI3EDuxbD9hyzmz/bbOz5jp/dfPNHhU3ixubDGx/8KLOW45+BQ0sDKl8CLHIAyOLBqh4I5IkQGQWjYBSMgpEOAGc9WnxxSAX3AAAAAElFTkSuQmCC","orcid":"","institution":"Xi'an Jiaotong University","correspondingAuthor":true,"prefix":"","firstName":"Junmin","middleName":"","lastName":"Liu","suffix":""}],"badges":[],"createdAt":"2026-04-02 04:38:25","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9298008/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9298008/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108006883,"identity":"f39300f1-cde0-4715-a25d-fdea5227a5be","added_by":"auto","created_at":"2026-04-28 12:57:47","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1244668,"visible":true,"origin":"","legend":"","description":"","filename":"DegradationGuidedPromptingandDynamicTaskCurriculumforAllinOneImageRestoration.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9298008/v1_covered_8374c60d-5b1a-42de-8256-8a19ecce7708.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Degradation-Guided Prompting and Dynamic Task Curriculum for All-in-One Image Restoration","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":"All-in-one image restoration, multiple degradation types, degradation-guided prompt learning, dynamic task curriculum","lastPublishedDoi":"10.21203/rs.3.rs-9298008/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9298008/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAll-in-one image restoration has emerged as a pivotal paradigm in low-level vision. Unlike conventional single-task approaches, this paradigm aims to learn a unified model capable of handling multiple degradation types within a single framework. However, most existing methods predominantly suffer from two critical limitations: i) cross-degradation knowledge contamination during joint optimization, and ii) catastrophic forgetting in sequential learning scenarios. To address these limitations, we introduce a Degradation-Guided Prompt Learning Network (DPLNet), a unified architecture that employs a plug-and-play prompt block to explicitly mine degradation-specific priors from the input, thereby injecting task-adaptive guidance into the restoration process. Complementing the architecture, we introduce the Dynamic Task Curriculum (DTC), a principled learning strategy that sequentially schedules multiple tasks within each training epoch. DTC quantifies per-instance hardness through an online uncertainty criterion and progressively enriches each task optimization by retaining the hard examples from previous tasks, yielding continual robustness gains. Extensive experiments demonstrate that DPLNet achieves state-of-the-art performance in both single-task and multi-task settings. And our proposed training strategy yields consistent improvements across existing architectures.\u003c/p\u003e","manuscriptTitle":"Degradation-Guided Prompting and Dynamic Task Curriculum for All-in-One Image Restoration","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-28 04:45:49","doi":"10.21203/rs.3.rs-9298008/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-18T22:12:33+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-28T22:48:19+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"328930042509819121091572876788255293309","date":"2026-04-20T10:40:21+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"178496410147554387001762634516238233144","date":"2026-04-20T08:34:34+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"296120393821695522281041191430467262775","date":"2026-04-20T06:55:20+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"132465502406942786200100343263876990482","date":"2026-04-20T05:21:51+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-20T04:13:30+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-07T03:14:36+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-02T10:40:28+00:00","index":"","fulltext":""},{"type":"submitted","content":"International Journal of Machine Learning and Cybernetics","date":"2026-04-02T04:36:44+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":"93804d43-28b1-4856-87ed-5c76152695e1","owner":[],"postedDate":"April 28th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-18T22:12:33+00:00","index":19,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-28T04:45:49+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-28 04:45:49","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9298008","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9298008","identity":"rs-9298008","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}