Domain-Partitioned Retrieval as a Hallucination Mitigation Strategy in Conversational RAG: The PC-RAG Architecture

Domain-Partitioned Retrieval as a Hallucination Mitigation Strategy in Conversational RAG: The PC-RAG Architecture | 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 Domain-Partitioned Retrieval as a Hallucination Mitigation Strategy in Conversational RAG: The PC-RAG Architecture Ruben Alejandro Jaime This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9334334/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 Hallucination in Retrieval-Augmented Generation (RAG) systems is commonly attributed to generative model limitations. We argue that the primary cause in enterprise document retrieval is context window contamination: single-pass retrieval aggregates chunks from semantically heterogeneous domains, forcing the synthesis model to reason over a noisy cross-domain context and producing coherent but factually unstable responses. We present PC-RAG (Pipeline-Conversational RAG), an architecture built on domain-partitioned retrieval: complex queries are decomposed into domain-isolated sub-tasks, each serviced by an independent retrieval pass. Within each partition, intra-domain multi-query expansion generates k=6 semantic variants, maximizing coverage without cross-domain interference. Evaluation on a 1,200-document production deployment (HR, legal, compliance) shows that domain partitioning alone reduces false-positive chunk inclusion by 51% and raises human-rated synthesis quality from 3.1 to 4.1 on a 5-point scale (two annotators, κ=0.79, Wilcoxon p<0.001). The full PC-RAG system achieves a synthesis score of 4.2, outperforming HyDE Gao et al. [2023] (3.6), standard multi-query retrieval (3.7), and RAPTOR Sarthi et al. [2024] (3.9). RAGAS automated evaluation corroborates human judgements (r=0.83, p<0.001). By contrast, upgrading the synthesis model alone yields only marginal gains (3.1 to 3.4, p=0.031), confirming the bottleneck is retrieval architecture, not model capacity. The pipeline further integrates conversational coreference resolution (97.4% accuracy) and a semantic reranker; median end-to-end latency is 90 s, with primary contributions accounting for only 12% of this cost. Retrieval-Augmented Generation Hallucination mitigation Domain partitioning Multi-query expansion Conversational RAG Enterprise document intelligence Context window contamination 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-9334334","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":620822419,"identity":"d7742424-4910-4a6b-a140-ce3e691ad6ed","order_by":0,"name":"Ruben Alejandro Jaime","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABFUlEQVRIie2PMUvEMBTHHxTapdL15LD9BMILBVFO+jGcUwLN0uNudCwUOukHkA5+BQ8hOAYyuPgFxOW6uAkdvUE0bcRBTF0F8wvkveH98s8DcDj+IgEAAnhDS/WBeOjkdkrxvinpqNBflK97IK+MbSeqPbne3S1WEUDXd+Upvz5TnU7J4sPqZ2WmfEouH/jJVQVslotiuXkqUCssPZKWGBUi7jUKUQLVilpuWjooMhcWJVFRT96Mwl5z8c5Jy/tJBVUI6WdKoVMkTebldApRPqYHDcf9GorjXDByMy/XkqJ9l/i+fiYvzQKj4II97kSWJC2/7fvzLLauD+DjWLyQmq+Ok2gdH2e3pgbm1aSanHY4HI5/yAc0NGCPFlvEwgAAAABJRU5ErkJggg==","orcid":"","institution":"National University of La Plata","correspondingAuthor":true,"prefix":"","firstName":"Ruben","middleName":"Alejandro","lastName":"Jaime","suffix":""}],"badges":[],"createdAt":"2026-04-06 13:24:42","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9334334/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9334334/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":107481145,"identity":"d2ba38b4-badc-4ee6-a137-0419b205d089","added_by":"auto","created_at":"2026-04-22 02:16:06","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":279968,"visible":true,"origin":"","legend":"","description":"","filename":"pcragsubmission.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9334334/v1_covered_239b91a7-f0e2-4fe6-997c-9e93a993901e.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Domain-Partitioned Retrieval as a Hallucination Mitigation Strategy in Conversational RAG: The PC-RAG Architecture","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"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":"Retrieval-Augmented Generation, Hallucination mitigation, Domain partitioning, Multi-query expansion, Conversational RAG, Enterprise document intelligence, Context window contamination","lastPublishedDoi":"10.21203/rs.3.rs-9334334/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9334334/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eHallucination in Retrieval-Augmented Generation (RAG) systems is commonly attributed to generative model limitations. We argue that the primary cause in enterprise document retrieval is context window contamination: single-pass retrieval aggregates chunks from semantically heterogeneous domains, forcing the synthesis model to reason over a noisy cross-domain context and producing coherent but factually unstable responses. We present PC-RAG (Pipeline-Conversational RAG), an architecture built on domain-partitioned retrieval: complex queries are decomposed into domain-isolated sub-tasks, each serviced by an independent retrieval pass. Within each partition, intra-domain multi-query expansion generates k=6 semantic variants, maximizing coverage without cross-domain interference. Evaluation on a 1,200-document production deployment (HR, legal, compliance) shows that domain partitioning alone reduces false-positive chunk inclusion by 51% and raises human-rated synthesis quality from 3.1 to 4.1 on a 5-point scale (two annotators, κ=0.79, Wilcoxon p\u0026lt;0.001). The full PC-RAG system achieves a synthesis score of 4.2, outperforming HyDE Gao et al. [2023] (3.6), standard multi-query retrieval (3.7), and RAPTOR Sarthi et al. [2024] (3.9). RAGAS automated evaluation corroborates human judgements (r=0.83, p\u0026lt;0.001). By contrast, upgrading the synthesis model alone yields only marginal gains (3.1 to 3.4, p=0.031), confirming the bottleneck is retrieval architecture, not model capacity. The pipeline further integrates conversational coreference resolution (97.4% accuracy) and a semantic reranker; median end-to-end latency is 90 s, with primary contributions accounting for only 12% of this cost.\u003c/p\u003e","manuscriptTitle":"Domain-Partitioned Retrieval as a Hallucination Mitigation Strategy in Conversational RAG: The PC-RAG Architecture","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-16 03:00:40","doi":"10.21203/rs.3.rs-9334334/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":"475ff465-9384-46c5-948e-312a081e4678","owner":[],"postedDate":"April 16th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-16T03:00:40+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-16 03:00:40","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9334334","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9334334","identity":"rs-9334334","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}