Diversity-enhanced Conversational Recommendation via Multi-agent Reinforcement Learning

Diversity-enhanced Conversational Recommendation via Multi-agent Reinforcement Learning | 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 Diversity-enhanced Conversational Recommendation via Multi-agent Reinforcement Learning Zihan Wang, Shi Feng, Daling Wang, Kaisong Song, Gang Wu, Yifei Zhang, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4692909/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 22 May, 2025 Read the published version in Knowledge and Information Systems → Version 1 posted 9 You are reading this latest preprint version Abstract Multi-round Conversational Recommendation (MRCR) system assists users in finding the items they need with the fewest dialogue rounds by inquiring about desired features or making tailored recommendations. Numerous models employ single-agent Reinforcement Learning (RL) to accomplish MRCR and improve recommendation accuracy. However, they overlook the diversity of conversational recommendations and primarily focus on popular features or items. It impacts the fair visibility of the items and results in an unbalanced user experience. We propose a diversity-enhanced conversational recommendation model (DECREC), which is built on our proposed multi-agent RL framework. Three agents col-laboratively determine the actions at each round of the MRCR and each agent autonomously explores and learns distinct facets of the task. Compared to a single agent, their collaboration fosters the exploration of a more extensive array of actions to improve diversity. Furthermore, we introduce a dynamic experience replay method that balances long-tail and head data ensuring each learning batch includes long-tail samples, keeping the model attentive to these less common but important data. Moreover, we integrate feature entropy into the feature value estimation process during training to encourage the model to explore a broader spectrum of features, thereby indirectly enhancing the diversity of recommendation results. Extensive experiments on four public datasets demonstrate that DECREC reduces bias in MRCR and achieves optimal recommendation diversity and accuracy. Our code is available at https://github.com/wzhwzhwzh0921/ DECREC. Recommendation systems Conversational Recommendation Multi-agent Reinforcement Learning Experience Replay Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 22 May, 2025 Read the published version in Knowledge and Information Systems → Version 1 posted Editorial decision: Revision requested 30 Mar, 2025 Reviews received at journal 28 Mar, 2025 Reviews received at journal 26 Mar, 2025 Reviewers agreed at journal 13 Mar, 2025 Reviewers agreed at journal 11 Mar, 2025 Reviewers invited by journal 15 Sep, 2024 Editor assigned by journal 18 Jul, 2024 Submission checks completed at journal 06 Jul, 2024 First submitted to journal 05 Jul, 2024 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-4692909","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":332389793,"identity":"8ee738e2-1844-403c-b86d-437751786927","order_by":0,"name":"Zihan Wang","email":"","orcid":"","institution":"Northeastern University","correspondingAuthor":false,"prefix":"","firstName":"Zihan","middleName":"","lastName":"Wang","suffix":""},{"id":332389794,"identity":"9a6ee1e8-27f1-4941-8c2f-63b72bc06d4f","order_by":1,"name":"Shi Feng","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8UlEQVRIiWNgGAWjYBAC9gYgwdggIYckloBfC88BiBZjkrUwJDYQr4W99/DLnzss0vv7Dx9gLqg4zMDPnmPA8HMHHi0859Ksec9I5M64kZbAPOPMYQbJnjcGjL1ncGuxl8gxM2Zsk8jdIMFjwMzbdpjB4EaOATNjGx5b5N+YGf5sk0g34D//gZn332EGe4JaJHiMH/C2SSQYMOQwMPM2AG2RIKSFJ8cM6B4JQ6BfDA7zHEvnkTjzrOBgLz4t7GeMP/5sq5Pn7z/88DFPjbUcf3vyxgc/8WgBAjYJGOsAyAwYAx9g/kBAwSgYBaNgFIx0AABmwEpJb5blOgAAAABJRU5ErkJggg==","orcid":"","institution":"Northeastern University","correspondingAuthor":true,"prefix":"","firstName":"Shi","middleName":"","lastName":"Feng","suffix":""},{"id":332389795,"identity":"cb3a41bd-7336-4831-b1fe-e4c1fc03fe99","order_by":2,"name":"Daling Wang","email":"","orcid":"","institution":"Northeastern University","correspondingAuthor":false,"prefix":"","firstName":"Daling","middleName":"","lastName":"Wang","suffix":""},{"id":332389796,"identity":"d6f6f6e5-c262-4dc5-97e2-f6330058a8c2","order_by":3,"name":"Kaisong Song","email":"","orcid":"","institution":"Alibaba","correspondingAuthor":false,"prefix":"","firstName":"Kaisong","middleName":"","lastName":"Song","suffix":""},{"id":332389797,"identity":"9f3557d6-7f58-43a1-a340-ab4d093c8b14","order_by":4,"name":"Gang Wu","email":"","orcid":"","institution":"Northeastern University","correspondingAuthor":false,"prefix":"","firstName":"Gang","middleName":"","lastName":"Wu","suffix":""},{"id":332389798,"identity":"ba03cf20-bb76-4de9-870f-f4bb313345aa","order_by":5,"name":"Yifei Zhang","email":"","orcid":"","institution":"Northeastern University","correspondingAuthor":false,"prefix":"","firstName":"Yifei","middleName":"","lastName":"Zhang","suffix":""},{"id":332389799,"identity":"a567c7a9-92ae-4b33-ae79-13e6ea64fed9","order_by":6,"name":"Han Zhao","email":"","orcid":"","institution":"Northeastern University","correspondingAuthor":false,"prefix":"","firstName":"Han","middleName":"","lastName":"Zhao","suffix":""},{"id":332389800,"identity":"362d3e6f-6471-4ae8-b775-2aff32758c38","order_by":7,"name":"Ge Yu","email":"","orcid":"","institution":"Northeastern University","correspondingAuthor":false,"prefix":"","firstName":"Ge","middleName":"","lastName":"Yu","suffix":""}],"badges":[],"createdAt":"2024-07-05 14:29:47","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4692909/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4692909/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1007/s10115-025-02455-w","type":"published","date":"2025-05-22T15:57:54+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":83460187,"identity":"1099b5d2-7b5e-4504-afd5-27b6ef03b9d0","added_by":"auto","created_at":"2025-05-26 16:11:44","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1270526,"visible":true,"origin":"","legend":"","description":"","filename":"KAISDECREC2.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4692909/v1_covered_41d8a2e7-10dc-4468-8b8f-645797483535.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Diversity-enhanced Conversational Recommendation via Multi-agent Reinforcement Learning","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":"knowledge-and-information-systems","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"kais","sideBox":"Learn more about [Knowledge and Information Systems](http://link.springer.com/journal/10115)","snPcode":"10115","submissionUrl":"https://submission.nature.com/new-submission/10115/3","title":"Knowledge and Information Systems","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Recommendation systems, Conversational Recommendation, Multi-agent Reinforcement Learning, Experience Replay","lastPublishedDoi":"10.21203/rs.3.rs-4692909/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4692909/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Multi-round Conversational Recommendation (MRCR) system assists users in finding the items they need with the fewest dialogue rounds by inquiring about desired features or making tailored recommendations. Numerous models employ single-agent Reinforcement Learning (RL) to accomplish MRCR and improve recommendation accuracy. However, they overlook the diversity of conversational recommendations and primarily focus on popular features or items. It impacts the fair visibility of the items and results in an unbalanced user experience. We propose a diversity-enhanced conversational recommendation model (DECREC), which is built on our proposed multi-agent RL framework. Three agents col-laboratively determine the actions at each round of the MRCR and each agent autonomously explores and learns distinct facets of the task. Compared to a single agent, their collaboration fosters the exploration of a more extensive array of actions to improve diversity. Furthermore, we introduce a dynamic experience replay method that balances long-tail and head data ensuring each learning batch includes long-tail samples, keeping the model attentive to these less common but important data. Moreover, we integrate feature entropy into the feature value estimation process during training to encourage the model to explore a broader spectrum of features, thereby indirectly enhancing the diversity of recommendation results. Extensive experiments on four public datasets demonstrate that DECREC reduces bias in MRCR and achieves optimal recommendation diversity and accuracy. Our code is available at https://github.com/wzhwzhwzh0921/ DECREC. ","manuscriptTitle":"Diversity-enhanced Conversational Recommendation via Multi-agent Reinforcement Learning","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-08-13 02:54:08","doi":"10.21203/rs.3.rs-4692909/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-03-30T07:55:44+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-03-28T14:53:51+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-03-26T22:08:28+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"280317539334891312320559629824297519342","date":"2025-03-14T03:36:17+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"36925964666706541855293313524270584325","date":"2025-03-11T09:39:23+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-09-15T11:34:23+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-07-19T02:34:47+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-07-06T05:58:16+00:00","index":"","fulltext":""},{"type":"submitted","content":"Knowledge and Information Systems","date":"2024-07-05T14:28:33+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"knowledge-and-information-systems","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"kais","sideBox":"Learn more about [Knowledge and Information Systems](http://link.springer.com/journal/10115)","snPcode":"10115","submissionUrl":"https://submission.nature.com/new-submission/10115/3","title":"Knowledge and Information Systems","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"d7ecf379-af8c-41f9-bb99-01121afb21c4","owner":[],"postedDate":"August 13th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-05-26T16:06:22+00:00","versionOfRecord":{"articleIdentity":"rs-4692909","link":"https://doi.org/10.1007/s10115-025-02455-w","journal":{"identity":"knowledge-and-information-systems","isVorOnly":false,"title":"Knowledge and Information Systems"},"publishedOn":"2025-05-22 15:57:54","publishedOnDateReadable":"May 22nd, 2025"},"versionCreatedAt":"2024-08-13 02:54:08","video":"","vorDoi":"10.1007/s10115-025-02455-w","vorDoiUrl":"https://doi.org/10.1007/s10115-025-02455-w","workflowStages":[]},"version":"v1","identity":"rs-4692909","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4692909","identity":"rs-4692909","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}