Task-Conditioned Multi-Task Learning for Actor-Centric Rhetorical Positioning in Political Discourse

Task-Conditioned Multi-Task Learning for Actor-Centric Rhetorical Positioning in Political Discourse | 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 Task-Conditioned Multi-Task Learning for Actor-Centric Rhetorical Positioning in Political Discourse Samantha Butcher, Beatriz De La Iglesia This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9022111/v2 This work is licensed under a CC BY 4.0 License Status: Posted Version 2 posted You are reading this latest preprint version Show more versions Abstract Automatically modelling how political actors are evaluated in discourse requires methods that move beyond general sentiment classification. Existing NLP techniques typically capture overall polarity but struggle to attribute evaluative language to the specific social entities referenced within a text. This paper introduces actor-centric rhetorical positioning, a token-level framework that links references to social actors with Positive, Neutral, or Negative evaluative labels. To model the dependency between entity identification and evaluation, we propose a Task-Conditioned Multi-Task Learning (TC-MTL) architecture in which rhetorical positioning is conditioned on entity detection. We compare this approach against Single-Task Learning (STL) and parallel multi-task baselines using DistilBERT, RoBERTa, and DeBERTa on a corpus of U.S. political discourse spanning rally speeches, formal addresses, and online political commentary. TC-MTL improves NER performance by approximately 2–4 F1 points over STL baselines. Explicit conditioning strengthens entity extraction while maintaining positioning performance, resulting in a more reliable framework for analysing actor-centred political discourse. Our approach enables the systematic detection of how actors are constructed and positioned in political speech, supporting large-scale analysis of rhetorical framing and identity-based polarisation. Artificial Intelligence and Machine Learning Other Political Science machine learning discourse analysis multi-task learning Full Text Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 2 posted You are reading this latest preprint version Show more versions 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-9022111","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":600946816,"identity":"76fc8236-58fc-4858-a7c7-94237171a171","order_by":0,"name":"Samantha Butcher","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABDklEQVRIiWNgGAWjYPCCAwwMEkCKsYGBgR8uyINPRwJYC2MDSItkA8laDA4Q0CLffjrxc+GPO/IM0j3mD37usMk3vpF88ANDjR2DwZkDWLUYnMndLD0j4Zlhg8wZw8beM2mW226kJUswHEtmMDjbgF0LQ+4GaZ6Ew4wNEjmGzYxthw3MbuQYSDCwHWAwOI/DYf1vN/8GarGHavlvYDwj//MPhn+4tTDcyN0GsiURquWAgYFEDpsEkIHbYTfebrPmSTuc3CZzrHBmb1uygcSZZ2YWiX3JPJI4vC/fn7v5No/NYdt+6eYNH3622Rnwtyc/vvHhm50c35kEHC6DAjYUXgKBiBwFo2AUjIJRgB8AANRQYrDaGcKgAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0009-0000-0041-6768","institution":"University of East Anglia","correspondingAuthor":true,"prefix":"","firstName":"Samantha","middleName":"","lastName":"Butcher","suffix":""},{"id":600946817,"identity":"7f62e418-5c4e-44c8-927d-9859c259727d","order_by":1,"name":"Beatriz De La Iglesia","email":"","orcid":"https://orcid.org/0000-0003-2675-5826","institution":"University of East Anglia","correspondingAuthor":false,"prefix":"","firstName":"Beatriz","middleName":"De La","lastName":"Iglesia","suffix":""}],"badges":[],"createdAt":"2026-03-03 16:01:40","currentVersionCode":2,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-9022111/v2","doiUrl":"https://doi.org/10.21203/rs.3.rs-9022111/v2","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":105035712,"identity":"c2625fda-5dd3-42ec-ae7f-6385a6a11b9c","added_by":"auto","created_at":"2026-03-20 07:26:29","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":589077,"visible":true,"origin":"","legend":"","description":"","filename":"JournalNCA6.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9022111/v2_covered_b1b2892e-4f75-4f64-8965-697b0d6a087b.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"Task-Conditioned Multi-Task Learning for Actor-Centric Rhetorical Positioning in Political Discourse","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"University of East Anglia","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":"machine learning, discourse analysis, multi-task learning","lastPublishedDoi":"10.21203/rs.3.rs-9022111/v2","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9022111/v2","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eAutomatically modelling how political actors are evaluated in discourse requires methods that move beyond general sentiment classification. Existing NLP techniques typically capture overall polarity but struggle to attribute evaluative language to the specific social entities referenced within a text.\u003c/p\u003e\n\u003cp\u003eThis paper introduces actor-centric rhetorical positioning, a token-level framework that links references to social actors with Positive, Neutral, or Negative evaluative labels. To model the dependency between entity identification and evaluation, we propose a Task-Conditioned Multi-Task Learning (TC-MTL) architecture in which rhetorical positioning is conditioned on entity detection. We compare this approach against Single-Task Learning (STL) and parallel multi-task baselines using DistilBERT, RoBERTa, and DeBERTa on a corpus of U.S. political discourse spanning rally speeches, formal addresses, and online political commentary.\u003c/p\u003e\n\u003cp\u003eTC-MTL improves NER performance by approximately 2–4 F1 points over STL baselines. \u0026nbsp;Explicit conditioning strengthens entity extraction while maintaining positioning performance, resulting in a more reliable framework for analysing actor-centred political discourse. Our approach enables the systematic detection of how actors are constructed and positioned in political speech, supporting large-scale analysis of rhetorical framing and identity-based polarisation.\u003c/p\u003e","manuscriptTitle":"Task-Conditioned Multi-Task Learning for Actor-Centric Rhetorical Positioning in Political Discourse","msid":"","msnumber":"","nonDraftVersions":[{"code":2,"date":"2026-03-19 17:26:29","doi":"10.21203/rs.3.rs-9022111/v2","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}},{"code":1,"date":"2026-03-05 03:04:42","doi":"10.21203/rs.3.rs-9022111/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":"cebd6742-e9de-477f-a205-b5e68b962a21","owner":[],"postedDate":"March 19th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":63952354,"name":"Artificial Intelligence and Machine Learning"},{"id":63952355,"name":"Other Political Science"}],"tags":[],"updatedAt":"2026-03-05T03:04:42+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-19 17:26:29","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v2","identity":"rs-9022111","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9022111","identity":"rs-9022111","version":["v2"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}