MD2PR: A Multi-level Distillation based Dense Passage Retrieval Model

MD2PR: A Multi-level Distillation based Dense Passage Retrieval Model | 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 MD2PR: A Multi-level Distillation based Dense Passage Retrieval Model Haifeng Li, Mo Hai, Dong Tang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6219315/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 Reranker and retriever are two important components in information retrieval. The retriever typically adopts a dual-encoder model, where queries and documents are separately input into two pre-trained models, and the vectors generated by the models are used for similarity calculation. The reranker often uses a cross-encoder model, where the concatenated query-document pairs are input into a pre-trained model to obtain word similarities. However, the dual-encoder model lacks interaction between queries and documents due to its independent encoding, while the cross-encoder model requires substantial computational cost for attention calculation, making it difficult to obtain real-time retrieval results. In this paper, we propose a dense retrieval model called MD2PR based on multi-level knowledge distillation, that is, the knowledge learned from the cross-encoder is distilled to the dual-encoder at both the sentence level and word level. Sentence-level distillation enhances the dual-encoder on capturing the themes and emotions of sentences. Word-level distillation improves the dual-encoder in analysis of word semantics and relationships. As a result, the dual-encoder can be used independently for subsequent encoding and retrieval, avoiding the significant computational cost associated with the participation of the cross-encoder. Furthermore, we propose a dynamic false negative filtering method, which updates the threshold during multiple training iterations to ensure the effective identification of false negatives and thus obtains a more comprehensive semantic representation space. The experimental results over two standard datasets show our MD2PR outperforms 14 baseline models in terms of MRR and Recall metrics. MD2PR Multi-level Knowledge Distillation Dual-Encoder Cross-Encoder Retriever Reranker Full Text Additional Declarations No competing interests reported. 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-6219315","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":439804423,"identity":"e9b0ee20-0a5c-48e5-b2ec-c7210fb772a0","order_by":0,"name":"Haifeng Li","email":"","orcid":"","institution":"Central University of Finance and Economics","correspondingAuthor":false,"prefix":"","firstName":"Haifeng","middleName":"","lastName":"Li","suffix":""},{"id":439804424,"identity":"27c0405d-e5b7-4342-8138-549c7ce44800","order_by":1,"name":"Mo Hai","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAxUlEQVRIiWNgGAWjYBACAwbGBgaGCgYGdiDFwEO8ljNA1QeI1wIEjG2kaDFnP9y64eM8u8Qe9gbGB2/bGOTNCWmx7ElsuzlzW3JiD88BZsO5bQyGOxsIOexAYttt3m0HEvdLJLBJ87YxJBgcIKTl/EOgljkHEnvkH7D/Jk7LDZAtDUAtEgxszERqedh2c8axZOMensRmyTnnJAw3EHZY+rMbH2rsZHvYDx/88KbMRp6gLUgAFKcMEsSrHwWjYBSMglGAGwAASfJD82Kl2dMAAAAASUVORK5CYII=","orcid":"","institution":"Central University of Finance and Economics","correspondingAuthor":true,"prefix":"","firstName":"Mo","middleName":"","lastName":"Hai","suffix":""},{"id":439804427,"identity":"5b88de93-1798-47f5-bc93-a79c38271087","order_by":2,"name":"Dong Tang","email":"","orcid":"","institution":"Central University of Finance and Economics","correspondingAuthor":false,"prefix":"","firstName":"Dong","middleName":"","lastName":"Tang","suffix":""}],"badges":[],"createdAt":"2025-03-13 10:53:29","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6219315/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6219315/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":94040982,"identity":"f0b3be60-5a51-4afb-b61d-ed82702b8150","added_by":"auto","created_at":"2025-10-21 18:31:22","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":397129,"visible":true,"origin":"","legend":"","description":"","filename":"www.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6219315/v1_covered_9a355ee8-1bef-42a2-9931-d7b542315631.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"MD2PR: A Multi-level Distillation based Dense Passage Retrieval Model","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"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":"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":"MD2PR, Multi-level, Knowledge Distillation, Dual-Encoder, Cross-Encoder, Retriever, Reranker","lastPublishedDoi":"10.21203/rs.3.rs-6219315/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6219315/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eReranker and retriever are two important components in information retrieval. The retriever typically adopts a dual-encoder model, where queries and documents are separately input into two pre-trained models, and the vectors generated by the models are used for similarity calculation. The reranker often uses a cross-encoder model, where the concatenated query-document pairs are input into a pre-trained model to obtain word similarities. However, the dual-encoder model lacks interaction between queries and documents due to its independent encoding, while the cross-encoder model requires substantial computational cost for attention calculation, making it difficult to obtain real-time retrieval results. In this paper, we propose a dense retrieval model called MD2PR based on multi-level knowledge distillation, that is, the knowledge learned from the cross-encoder is distilled to the dual-encoder at both the sentence level and word level. Sentence-level distillation enhances the dual-encoder on capturing the themes and emotions of sentences. Word-level distillation improves the dual-encoder in analysis of word semantics and relationships. As a result, the dual-encoder can be used independently for subsequent encoding and retrieval, avoiding the significant computational cost associated with the participation of the cross-encoder. Furthermore, we propose a dynamic false negative filtering method, which updates the threshold during multiple training iterations to ensure the effective identification of false negatives and thus obtains a more comprehensive semantic representation space. The experimental results over two standard datasets show our MD2PR outperforms 14 baseline models in terms of MRR and Recall metrics.\u003c/p\u003e","manuscriptTitle":"MD2PR: A Multi-level Distillation based Dense Passage Retrieval Model","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-17 06:05:57","doi":"10.21203/rs.3.rs-6219315/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":"86955f20-36b4-4876-9fcf-675cfadaa997","owner":[],"postedDate":"April 17th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-10-21T18:23:15+00:00","versionOfRecord":[],"versionCreatedAt":"2025-04-17 06:05:57","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6219315","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6219315","identity":"rs-6219315","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}