Retrieval-Augmented Generation for Natural Language Processing: A Survey

Retrieval-Augmented Generation for Natural Language Processing: A Survey | 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 Retrieval-Augmented Generation for Natural Language Processing: A Survey Shangyu Wu, Ying Xiong, Yufei Cui, Haolun Wu, Can Chen, Ye Yuan, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6959723/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 15 You are reading this latest preprint version Abstract Large language models (LLMs) have demonstrated great success in various fields, benefiting from their huge amount of parameters that store knowledge.However, LLMs still suffer from several key issues, such as hallucination problems, knowledge update issues, and lacking domain-specific expertise.The appearance of retrieval-augmented generation (RAG), which leverages an external knowledge database to augment LLMs, makes up those drawbacks of LLMs.This paper reviews all significant techniques of RAG, especially in the retriever and the retrieval fusions.Besides, tutorial codes are provided for implementing the representative techniques in RAG.This paper further discusses the RAG update, including RAG with/without knowledge update.Then, we introduce RAG evaluation and benchmarking, as well as the application of RAG in representative NLP tasks and industrial scenarios.Finally, this paper discusses RAG's future directions and challenges for promoting this field's development. Retrieval-augmented generation natural language processing vector database large language model Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 17 Jan, 2026 Reviews received at journal 22 Dec, 2025 Reviews received at journal 19 Dec, 2025 Reviews received at journal 04 Dec, 2025 Reviews received at journal 17 Nov, 2025 Reviewers agreed at journal 04 Nov, 2025 Reviewers agreed at journal 04 Nov, 2025 Reviewers agreed at journal 04 Nov, 2025 Reviewers agreed at journal 04 Nov, 2025 Reviewers agreed at journal 29 Sep, 2025 Reviewers agreed at journal 25 Sep, 2025 Reviewers invited by journal 15 Aug, 2025 Editor assigned by journal 27 Jun, 2025 Submission checks completed at journal 23 Jun, 2025 First submitted to journal 23 Jun, 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-6959723","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":501233694,"identity":"e98c1d13-46bb-4263-a715-762b598eacb6","order_by":0,"name":"Shangyu Wu","email":"","orcid":"","institution":"City University of Hong Kong","correspondingAuthor":false,"prefix":"","firstName":"Shangyu","middleName":"","lastName":"Wu","suffix":""},{"id":501233695,"identity":"33853460-6508-4117-925d-6e78e9d95c52","order_by":1,"name":"Ying Xiong","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABB0lEQVRIiWNgGAWjYDACZjApAWQwH3zwoQIkwtxApBb2tmTDGWdAIowEtMABzxkzad42EIuAFoPjzMcefm2zyJOPyDGQ5p1XG83fDtTyo2IbTi2SzWzpxrJtEsWGN9IKDOduO5474zBjA2PPmds4tfAz85hJS7ZJJG6ckbwh4e22Y7kNQC3MjG24tbAhtCQYHOCdcyx3PiEtIFskPwK1zOc5YtjI21CTu4GQFqBf0qQZzkkkbgAGMuOMYwdyNwK1HMTnF4Pzh49J/iirS5zfzHz8x4eautx55w8ffPCjArcWEGDmZQPqPQBmHwaTB/CqBwLGH38YGOQbwOw6QopHwSgYBaNgBAIAi25c/MmlzYwAAAAASUVORK5CYII=","orcid":"","institution":"Mohamed bin Zayed University of Artificial Intelligence","correspondingAuthor":true,"prefix":"","firstName":"Ying","middleName":"","lastName":"Xiong","suffix":""},{"id":501233696,"identity":"92345318-d888-4848-9314-1465a3ffd906","order_by":2,"name":"Yufei Cui","email":"","orcid":"","institution":"McGill University","correspondingAuthor":false,"prefix":"","firstName":"Yufei","middleName":"","lastName":"Cui","suffix":""},{"id":501233697,"identity":"66b1fc43-90f3-4385-804d-402237c383bd","order_by":3,"name":"Haolun Wu","email":"","orcid":"","institution":"McGill University","correspondingAuthor":false,"prefix":"","firstName":"Haolun","middleName":"","lastName":"Wu","suffix":""},{"id":501233698,"identity":"913404d8-2cb9-444c-a85a-0dc673df3bb5","order_by":4,"name":"Can Chen","email":"","orcid":"","institution":"McGill University","correspondingAuthor":false,"prefix":"","firstName":"Can","middleName":"","lastName":"Chen","suffix":""},{"id":501233699,"identity":"9587e3d4-5adc-44a1-aa43-ff6cbb39e5e2","order_by":5,"name":"Ye Yuan","email":"","orcid":"","institution":"McGill University","correspondingAuthor":false,"prefix":"","firstName":"Ye","middleName":"","lastName":"Yuan","suffix":""},{"id":501233700,"identity":"3fd8daff-5657-4d1d-b8ea-ab3beb40c02f","order_by":6,"name":"Lianming Huang","email":"","orcid":"","institution":"City University of Hong Kong","correspondingAuthor":false,"prefix":"","firstName":"Lianming","middleName":"","lastName":"Huang","suffix":""},{"id":501233701,"identity":"40587309-a858-438e-b637-273fb2cf09b8","order_by":7,"name":"Xue Liu","email":"","orcid":"","institution":"Mohamed bin Zayed University of Artificial Intelligence","correspondingAuthor":false,"prefix":"","firstName":"Xue","middleName":"","lastName":"Liu","suffix":""},{"id":501233702,"identity":"dc0e8602-b2bc-45c1-820e-1d8349f4f611","order_by":8,"name":"Tei-Wei Kuo","email":"","orcid":"","institution":"National Taiwan University Hospital","correspondingAuthor":false,"prefix":"","firstName":"Tei-Wei","middleName":"","lastName":"Kuo","suffix":""},{"id":501233703,"identity":"ad3d9628-412f-4e8d-a61a-bd00422d21a3","order_by":9,"name":"Nan Guan","email":"","orcid":"","institution":"City University of Hong Kong","correspondingAuthor":false,"prefix":"","firstName":"Nan","middleName":"","lastName":"Guan","suffix":""},{"id":501233704,"identity":"60ffeabc-6eb8-4ebd-9d15-3302288bac49","order_by":10,"name":"Chun Xue","email":"","orcid":"","institution":"Mohamed bin Zayed University of Artificial Intelligence","correspondingAuthor":false,"prefix":"","firstName":"Chun","middleName":"","lastName":"Xue","suffix":""}],"badges":[],"createdAt":"2025-06-23 20:38:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6959723/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6959723/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":89685600,"identity":"57a8b4ba-c8a1-48c5-8105-cc1033689014","added_by":"auto","created_at":"2025-08-22 15:26:09","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":988808,"visible":true,"origin":"","legend":"","description":"","filename":"SNRAG.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6959723/v1_covered_1779c021-2e05-45da-8b0b-16c38888abfb.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Retrieval-Augmented Generation for Natural Language Processing: A Survey","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":"artificial-intelligence-review","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"aire","sideBox":"Learn more about [Artificial Intelligence Review](http://link.springer.com/journal/10462)","snPcode":"10462","submissionUrl":"https://submission.nature.com/new-submission/10462/3","title":"Artificial Intelligence Review","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Retrieval-augmented generation, natural language processing, vector database, large language model","lastPublishedDoi":"10.21203/rs.3.rs-6959723/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6959723/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Large language models (LLMs) have demonstrated great success in various fields, benefiting from their huge amount of parameters that store knowledge.However, LLMs still suffer from several key issues, such as hallucination problems, knowledge update issues, and lacking domain-specific expertise.The appearance of retrieval-augmented generation (RAG), which leverages an external knowledge database to augment LLMs, makes up those drawbacks of LLMs.This paper reviews all significant techniques of RAG, especially in the retriever and the retrieval fusions.Besides, tutorial codes are provided for implementing the representative techniques in RAG.This paper further discusses the RAG update, including RAG with/without knowledge update.Then, we introduce RAG evaluation and benchmarking, as well as the application of RAG in representative NLP tasks and industrial scenarios.Finally, this paper discusses RAG's future directions and challenges for promoting this field's development.","manuscriptTitle":"Retrieval-Augmented Generation for Natural Language Processing: A Survey","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-22 15:18:03","doi":"10.21203/rs.3.rs-6959723/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-01-17T12:42:57+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-23T00:08:38+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-19T11:43:26+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-05T02:30:44+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-17T19:28:26+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"66745325872474403536583050242072533062","date":"2025-11-04T20:38:11+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"158773247823164871494899312620006066236","date":"2025-11-04T19:35:45+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"22793707194479322004275196452295069940","date":"2025-11-04T18:15:15+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"169141298649782398302687414225659362938","date":"2025-11-04T17:00:26+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"107644549431573124852477367829279064653","date":"2025-09-29T11:45:46+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"6755634751227961276643793229785551684","date":"2025-09-25T17:28:07+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-15T08:26:08+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-06-27T07:53:51+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-06-24T01:14:04+00:00","index":"","fulltext":""},{"type":"submitted","content":"Artificial Intelligence Review","date":"2025-06-23T20:32:02+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"artificial-intelligence-review","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"aire","sideBox":"Learn more about [Artificial Intelligence Review](http://link.springer.com/journal/10462)","snPcode":"10462","submissionUrl":"https://submission.nature.com/new-submission/10462/3","title":"Artificial Intelligence Review","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"93035a5f-2eca-4713-b019-e5946715db5d","owner":[],"postedDate":"August 22nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-05-19T11:25:42+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-22 15:18:03","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6959723","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6959723","identity":"rs-6959723","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}