CAREFL: Context-Aware Recognition of Emotions with Federated Learning

CAREFL: Context-Aware Recognition of Emotions with Federated 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 Article CAREFL: Context-Aware Recognition of Emotions with Federated Learning Jose Alejandro Lopez Quel, Carlo Marcelo Revoredo da Silva, Alexander Fabisch, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8960272/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Recent advances in vision-language models (VLMs) have significantly improved visual emotion recognition using multimodal contextual reasoning. However, their deployment remains limited by heavy computational demands and centralized data requirements, limiting accessibility for privacy-sensitive or resource-limited settings. This work introduces CAREFL, a context-aware and federated framework that combines the reasoning capabilities of large foundation VLMs with the efficiency of compact, quantized models. In CAREFL, rich textual context is first generated from a powerful VLM and then used to guide the local fine-tuning of a small model under a federated learning setup. This approach enables distributed adaptation of multimodal emotion understanding while minimizing communication cost and preserving data privacy. Experiments on EMOTIC and CAER-S datasets demonstrate that CAREFL achieves up to 40.20% recall and 55.40% F1-score, surpassing heavier centralized baselines while operating within a 6 GB memory budget. The results suggest that leveraging contextual reasoning from foundation models can unlock scalable, privacy-preserving emotion recognition across heterogeneous edge environments. Physical sciences/Engineering Physical sciences/Mathematics and computing Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 28 Apr, 2026 Reviewers agreed at journal 15 Mar, 2026 Reviewers invited by journal 04 Mar, 2026 Editor invited by journal 04 Mar, 2026 Editor assigned by journal 26 Feb, 2026 Submission checks completed at journal 26 Feb, 2026 First submitted to journal 24 Feb, 2026 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-8960272","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":602669708,"identity":"04e54454-3dfb-4c3b-8bff-e26165e64a4f","order_by":0,"name":"Jose Alejandro Lopez Quel","email":"","orcid":"","institution":"Universidade de Pernambuco","correspondingAuthor":false,"prefix":"","firstName":"Jose","middleName":"Alejandro Lopez","lastName":"Quel","suffix":""},{"id":602669709,"identity":"716d9206-b870-4f24-bdc1-f664656b2cda","order_by":1,"name":"Carlo Marcelo Revoredo da Silva","email":"","orcid":"","institution":"Universidade de Pernambuco","correspondingAuthor":false,"prefix":"","firstName":"Carlo","middleName":"Marcelo Revoredo da","lastName":"Silva","suffix":""},{"id":602669710,"identity":"be840dbe-8f66-4931-866d-3f14ade95439","order_by":2,"name":"Alexander Fabisch","email":"","orcid":"","institution":"German Research Centre for Artificial Intelligence","correspondingAuthor":false,"prefix":"","firstName":"Alexander","middleName":"","lastName":"Fabisch","suffix":""},{"id":602669711,"identity":"bb2e471a-d1ee-4ff1-9e93-df941e9c3958","order_by":3,"name":"Bruno José Torres Fernandes","email":"data:image/png;base64,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","orcid":"","institution":"Leibniz University Hannover","correspondingAuthor":true,"prefix":"","firstName":"Bruno","middleName":"José Torres","lastName":"Fernandes","suffix":""}],"badges":[],"createdAt":"2026-02-24 18:23:51","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8960272/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8960272/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104404955,"identity":"7828028b-f6ff-4a50-81b9-369650c22283","added_by":"auto","created_at":"2026-03-11 12:21:26","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1774176,"visible":true,"origin":"","legend":"","description":"","filename":"ScientificReportsJALQ2026.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8960272/v1_covered_2a768db0-1d50-4498-be64-888e7a88ce18.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"CAREFL: Context-Aware Recognition of Emotions with Federated Learning","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":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8960272/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8960272/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Recent advances in vision-language models (VLMs) have significantly improved visual emotion recognition using multimodal contextual reasoning. However, their deployment remains limited by heavy computational demands and centralized data requirements, limiting accessibility for privacy-sensitive or resource-limited settings. This work introduces CAREFL, a context-aware and federated framework that combines the reasoning capabilities of large foundation VLMs with the efficiency of compact, quantized models. In CAREFL, rich textual context is first generated from a powerful VLM and then used to guide the local fine-tuning of a small model under a federated learning setup. This approach enables distributed adaptation of multimodal emotion understanding while minimizing communication cost and preserving data privacy. Experiments on EMOTIC and CAER-S datasets demonstrate that CAREFL achieves up to 40.20\\% recall and 55.40\\% F1-score, surpassing heavier centralized baselines while operating within a 6 GB memory budget. The results suggest that leveraging contextual reasoning from foundation models can unlock scalable, privacy-preserving emotion recognition across heterogeneous edge environments.","manuscriptTitle":"CAREFL: Context-Aware Recognition of Emotions with Federated Learning","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-09 21:27:38","doi":"10.21203/rs.3.rs-8960272/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-04-28T13:42:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"146817435956532654508891779588191732382","date":"2026-03-15T22:38:41+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-05T00:13:56+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2026-03-04T20:56:00+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-26T09:49:40+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-26T09:46:06+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2026-02-24T18:11:15+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"d464c67f-3f1c-4bb3-9202-b2f69cc7c886","owner":[],"postedDate":"March 9th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":64129658,"name":"Physical sciences/Engineering"},{"id":64129659,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2026-03-09T21:27:38+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-09 21:27:38","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8960272","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8960272","identity":"rs-8960272","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}