Dynamic RAG: A Framework for Synergistic Collaboration Between Large and Small Models in Cloud-Edge Intelligence

Dynamic RAG: A Framework for Synergistic Collaboration Between Large and Small Models in Cloud-Edge Intelligence | 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 Dynamic RAG: A Framework for Synergistic Collaboration Between Large and Small Models in Cloud-Edge Intelligence XinLu Zhang, JiaYun Fang, ZhengChao Ding, LiQiang Tang, JianJun Jiang, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8244994/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 The synergy between powerful, cloud-hosted large models and efficient, on-device small models presents a new frontier for mobile artificial intelligence. However, enabling real-time, knowledge-intensive interactions, such as querying physical objects in the user's environment, is severely hampered by the high latency of cloud-centric approaches and the limited reasoning capabilities of edge-only models. This paper introduces a novel synergistic cloud-edge intelligence framework designed to bridge this critical gap. Our framework implements a semantic partitioning of labor: a lightweight edge client performs real-time perception to generate compact multimodal features, while an intelligent trigger decides when to offload a structured, minimal-data payload to a powerful cloud-based reasoning engine for definitive entity linking and knowledge-grounded question answering. We validate our approach on SynergyQA, a new benchmark for this task. Experimental results show that our framework reduces end-to-end latency by over 80% and data transmission by over 98% compared to naive offloading strategies, while achieving question-answering performance that is competitive with state-of-the-art cloud-only models. Our work provides an efficient and robust blueprint for deploying the next generation of interactive, context-aware AI on mobile and edge devices. Cloud-Edge Collaboration Multimodal Large Models Edge Intelligence Multimodal Entity Linking Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 06 May, 2026 Reviewers agreed at journal 23 Mar, 2026 Reviews received at journal 23 Feb, 2026 Reviewers agreed at journal 16 Feb, 2026 Reviewers agreed at journal 09 Feb, 2026 Reviewers agreed at journal 09 Feb, 2026 Reviewers invited by journal 08 Feb, 2026 Editor assigned by journal 05 Dec, 2025 Submission checks completed at journal 04 Dec, 2025 First submitted to journal 30 Nov, 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-8244994","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":589016255,"identity":"2f3a2771-13e4-493f-ba84-e6e8f4763bbb","order_by":0,"name":"XinLu Zhang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA4ElEQVRIiWNgGAWjYDACCSjJB6I+GNjYEa+FDUgyzihISyZWCwNYCzPPh0OMDYR0yM9ufvbwa5tFHptEjpm0jcEBZgb2w0c34NPCOOeYubHMGYlisJYcgzt8DDxpaTfwaWGWSDCTlqiQSGyTyN0G1PKMmUGCxwyvFjaJ9G/SEgZQLRYGhxkbCGnhAbpH8gPMFgZitEhI5JRJM5wBauF5/9myxyAtmY2QX+RnpG+T/NlWl9jPnpZ448cfGzt+9sPH8GoBAWYeCM0CjiM2QspBgPEHVOsHYlSPglEwCkbByAMA+n9Blz49wEcAAAAASUVORK5CYII=","orcid":"","institution":"China Electronics Technology Group Corporation (China)","correspondingAuthor":true,"prefix":"","firstName":"XinLu","middleName":"","lastName":"Zhang","suffix":""},{"id":589016258,"identity":"535fa9b5-0184-413f-847b-987c897e21b9","order_by":1,"name":"JiaYun Fang","email":"","orcid":"","institution":"China Electronics Technology Group Corporation (China)","correspondingAuthor":false,"prefix":"","firstName":"JiaYun","middleName":"","lastName":"Fang","suffix":""},{"id":589016260,"identity":"3b68d09b-2a38-4fac-a6a8-255c964690d6","order_by":2,"name":"ZhengChao Ding","email":"","orcid":"","institution":"China Electronics Technology Group Corporation (China)","correspondingAuthor":false,"prefix":"","firstName":"ZhengChao","middleName":"","lastName":"Ding","suffix":""},{"id":589016261,"identity":"d8e558b3-7eec-4fb1-8b9d-71448e884fd9","order_by":3,"name":"LiQiang Tang","email":"","orcid":"","institution":"China Electronics Technology Group Corporation (China)","correspondingAuthor":false,"prefix":"","firstName":"LiQiang","middleName":"","lastName":"Tang","suffix":""},{"id":589016263,"identity":"383970ed-368c-4c07-9519-dc13caadd8f0","order_by":4,"name":"JianJun Jiang","email":"","orcid":"","institution":"China Electronics Technology Group Corporation (China)","correspondingAuthor":false,"prefix":"","firstName":"JianJun","middleName":"","lastName":"Jiang","suffix":""},{"id":589016264,"identity":"6c99fa0b-5410-40f7-9c76-e92678f719cf","order_by":5,"name":"Jun Yin","email":"","orcid":"","institution":"China Electronics Technology Group Corporation (China)","correspondingAuthor":false,"prefix":"","firstName":"Jun","middleName":"","lastName":"Yin","suffix":""},{"id":589016265,"identity":"320dc81c-037b-4740-9db8-f2ef8f37ffa1","order_by":6,"name":"LiEn Qian","email":"","orcid":"","institution":"China Electronics Technology Group Corporation (China)","correspondingAuthor":false,"prefix":"","firstName":"LiEn","middleName":"","lastName":"Qian","suffix":""}],"badges":[],"createdAt":"2025-12-01 01:23:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8244994/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8244994/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102746747,"identity":"26db9343-9a2d-4874-a80e-22688bd92d52","added_by":"auto","created_at":"2026-02-16 09:00:58","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":591855,"visible":true,"origin":"","legend":"","description":"","filename":"DynamicRAGAFrameworkforSynergisticCollaborationBetweenLargeandSmallModelsinCloudEdgeIntelligence.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8244994/v1_covered_4c8331cc-7d2a-43eb-88d0-33d0c3413f0f.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Dynamic RAG: A Framework for Synergistic Collaboration Between Large and Small Models in Cloud-Edge Intelligence","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":"journal-of-cloud-computing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"clco","sideBox":"Learn more about [Journal of Cloud Computing](http://journalofcloudcomputing.springeropen.com)","snPcode":"13677","submissionUrl":"https://submission.nature.com/new-submission/13677/3","title":"Journal of Cloud Computing","twitterHandle":"@SpringerOpen","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Cloud-Edge Collaboration, Multimodal Large Models, Edge Intelligence, Multimodal Entity Linking","lastPublishedDoi":"10.21203/rs.3.rs-8244994/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8244994/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"The synergy between powerful, cloud-hosted large models and efficient, on-device small models presents a new frontier for mobile artificial intelligence. \nHowever, enabling real-time, knowledge-intensive interactions, such as querying physical objects in the user's environment, is severely hampered by the high latency of cloud-centric approaches and the limited reasoning capabilities of edge-only models. \nThis paper introduces a novel synergistic cloud-edge intelligence framework designed to bridge this critical gap.\nOur framework implements a semantic partitioning of labor: a lightweight edge client performs real-time perception to generate compact multimodal features, while an intelligent trigger decides when to offload a structured, minimal-data payload to a powerful cloud-based reasoning engine for definitive entity linking and knowledge-grounded question answering. \nWe validate our approach on SynergyQA, a new benchmark for this task. Experimental results show that our framework reduces end-to-end latency by over 80\\% and data transmission by over 98\\% compared to naive offloading strategies, while achieving question-answering performance that is competitive with state-of-the-art cloud-only models. \nOur work provides an efficient and robust blueprint for deploying the next generation of interactive, context-aware AI on mobile and edge devices.","manuscriptTitle":"Dynamic RAG: A Framework for Synergistic Collaboration Between Large and Small Models in Cloud-Edge Intelligence","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-12 17:12:37","doi":"10.21203/rs.3.rs-8244994/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"81888372503287619212299107287196015600","date":"2026-05-06T13:20:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"98227860843831829673404189195933625405","date":"2026-03-23T13:54:54+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-23T11:31:26+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"31570800148108866721430219914697993848","date":"2026-02-16T09:11:25+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"258560340419206554565491922218107587584","date":"2026-02-09T09:43:18+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"131274185018697352298829408654827510900","date":"2026-02-09T07:57:57+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-09T00:41:49+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-12-05T08:20:28+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-12-04T17:32:00+00:00","index":"","fulltext":""},{"type":"submitted","content":"Journal of Cloud Computing","date":"2025-12-01T01:13:45+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"journal-of-cloud-computing","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"clco","sideBox":"Learn more about [Journal of Cloud Computing](http://journalofcloudcomputing.springeropen.com)","snPcode":"13677","submissionUrl":"https://submission.nature.com/new-submission/13677/3","title":"Journal of Cloud Computing","twitterHandle":"@SpringerOpen","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"9a976f35-6993-484e-87a8-7af862cc4525","owner":[],"postedDate":"February 12th, 2026","published":true,"recentEditorialEvents":[{"type":"reviewerAgreed","content":"81888372503287619212299107287196015600","date":"2026-05-06T13:20:12+00:00","index":57,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-02-12T17:12:37+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-12 17:12:37","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8244994","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8244994","identity":"rs-8244994","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}