Confidence-Aware Dual-Graph Calibration for Noise-Robust Relation Extraction in Low-Resource Scenarios

Confidence-Aware Dual-Graph Calibration for Noise-Robust Relation Extraction in Low-Resource Scenarios | 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 Confidence-Aware Dual-Graph Calibration for Noise-Robust Relation Extraction in Low-Resource Scenarios Jiafeng Suo, Jian Fang, He wang, Yang Xiao This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8872718/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 6 You are reading this latest preprint version Abstract Relation Extraction (RE) models often struggle in real-world settings with data scarcity and noisy supervision, where limited training samples amplify overfitting to spurious correlations and unreliable evidence. Graph-based approaches can exploit structural cues, yet they face a key trade-off: syntactic dependency graphs provide stable but potentially incomplete (and parser-dependent) structures, whereas learnable semantic affinity graphs are more flexible but prone to noisy connections, especially in low-resource regimes.To address these challenges, we propose Confidence-Aware Dual-Graph Calibration (CDGC), a unified framework for noise-robust RE. CDGC constructs two complementary views over the same sentence: a fixed syntactic dependency graph for structural guidance and a learnable semantic affinity graph for contextual interactions. We introduce a Structure-Guided Semantic Calibration module that injects syntactic priors to regularize dense semantic affinities and suppress spurious semantic edges. Furthermore, we devise a confidence-aware denoising strategy that measures structural--semantic predictive inconsistency via Jensen--Shannon divergence, and uses it to reweight unreliable instances and perform confidence-adaptive nucleus pruning on the semantic graph, without requiring additional clean supervision for denoising.Extensive experiments on TACRED, FewRel, and SemEval-2010 Task 8 under diverse low-resource and noisy-label protocols show that CDGC consistently outperforms strong structural and noise-robust baselines; for example, on TACRED with 10% training data and 30% label noise, CDGC improves micro-F1 from 48.7 to 53.2 (+4.5) over RoBERTa. Beyond extraction performance, we explicitly evaluate probabilistic calibration and show that CDGC yields substantially more reliable confidence estimates, reducing ECE from 0.142 to 0.051 and Brier score from 0.354 to 0.185 on the same setting. Additionally, we evaluate on DocRED to demonstrate that the dual-graph framework generalizes to document-level reasoning under low-resource clean supervision. relation extraction noisy supervision graph neural networks calibration Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 31 Mar, 2026 Reviewers agreed at journal 19 Feb, 2026 Reviewers invited by journal 19 Feb, 2026 Editor assigned by journal 17 Feb, 2026 Submission checks completed at journal 17 Feb, 2026 First submitted to journal 13 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-8872718","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":595280443,"identity":"0a854ebb-14f5-4ed8-9390-4ee3aa3daf06","order_by":0,"name":"Jiafeng Suo","email":"","orcid":"","institution":"Jilin Communications Polytechnic","correspondingAuthor":false,"prefix":"","firstName":"Jiafeng","middleName":"","lastName":"Suo","suffix":""},{"id":595280445,"identity":"e1a73e56-56b2-468a-897c-7cb2ba88e38c","order_by":1,"name":"Jian Fang","email":"","orcid":"","institution":"Jilin Communications Polytechnic","correspondingAuthor":false,"prefix":"","firstName":"Jian","middleName":"","lastName":"Fang","suffix":""},{"id":595280448,"identity":"8f2322ed-03bc-4ac5-9f44-7d17128717a4","order_by":2,"name":"He wang","email":"","orcid":"","institution":"Jilin Communications Polytechnic","correspondingAuthor":false,"prefix":"","firstName":"He","middleName":"","lastName":"wang","suffix":""},{"id":595280453,"identity":"c571193e-e27b-42f8-93d0-daad844750c8","order_by":3,"name":"Yang Xiao","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3ElEQVRIie2PvQrCMBSFbwi0S2PWgKCvEHBUfBahoKtT57joUpwrPoVvELlQHEq7CjoUfAHBpfgDRups6iaYbzjJcD44F8Dh+Em2pYkBbYE2r9dIQWliTL0vFPpS8F1uonRnSE7Ta+F7JJNwjhD4Wn1WpEbaWy0PZlgmSZIjiKO2KFCkbRbXCmVzBClGlmEK/RuL81p5NFFAo0eDStcKaaK8bmkzFRolnW7jfBKIvW1YguQS3IchF7gpq6jf4YltmNAAZA4hiJ00XwgsfQNXJu4wBL4o7W2Hw+H4S55vRkQKY7kLEwAAAABJRU5ErkJggg==","orcid":"","institution":"Jilin Communications Polytechnic","correspondingAuthor":true,"prefix":"","firstName":"Yang","middleName":"","lastName":"Xiao","suffix":""}],"badges":[],"createdAt":"2026-02-13 14:08:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8872718/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8872718/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":103506294,"identity":"5728723f-4734-4545-8580-15aa15153685","added_by":"auto","created_at":"2026-02-26 13:34:55","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":992394,"visible":true,"origin":"","legend":"","description":"","filename":"upload.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8872718/v1_covered_a5706c34-ceeb-4290-ad45-8508a7348184.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Confidence-Aware Dual-Graph Calibration for Noise-Robust Relation Extraction in Low-Resource Scenarios","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":"neural-processing-letters","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"nepl","sideBox":"Learn more about [Neural Processing Letters](http://link.springer.com/journal/11063)","snPcode":"11063","submissionUrl":"https://submission.nature.com/new-submission/11063/3","title":"Neural Processing Letters","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"relation extraction, noisy supervision, graph neural networks, calibration","lastPublishedDoi":"10.21203/rs.3.rs-8872718/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8872718/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eRelation Extraction (RE) models often struggle in real-world settings with data scarcity and noisy supervision, where limited training samples amplify overfitting to spurious correlations and unreliable evidence. Graph-based approaches can exploit structural cues, yet they face a key trade-off: syntactic dependency graphs provide stable but potentially incomplete (and parser-dependent) structures, whereas learnable semantic affinity graphs are more flexible but prone to noisy connections, especially in low-resource regimes.To address these challenges, we propose Confidence-Aware Dual-Graph Calibration (CDGC), a unified framework for noise-robust RE. CDGC constructs two complementary views over the same sentence: a fixed syntactic dependency graph for structural guidance and a learnable semantic affinity graph for contextual interactions. We introduce a Structure-Guided Semantic Calibration module that injects syntactic priors to regularize dense semantic affinities and suppress spurious semantic edges. Furthermore, we devise a confidence-aware denoising strategy that measures structural--semantic predictive inconsistency via Jensen--Shannon divergence, and uses it to reweight unreliable instances and perform confidence-adaptive nucleus pruning on the semantic graph, without requiring additional clean supervision for denoising.Extensive experiments on TACRED, FewRel, and SemEval-2010 Task 8 under diverse low-resource and noisy-label protocols show that CDGC consistently outperforms strong structural and noise-robust baselines; for example, on TACRED with 10% training data and 30% label noise, CDGC improves micro-F1 from 48.7 to 53.2 (+4.5) over RoBERTa. Beyond extraction performance, we explicitly evaluate probabilistic calibration and show that CDGC yields substantially more reliable confidence estimates, reducing ECE from 0.142 to 0.051 and Brier score from 0.354 to 0.185 on the same setting. Additionally, we evaluate on DocRED to demonstrate that the dual-graph framework generalizes to document-level reasoning under low-resource clean supervision.\u003c/p\u003e","manuscriptTitle":"Confidence-Aware Dual-Graph Calibration for Noise-Robust Relation Extraction in Low-Resource Scenarios","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-24 15:00:06","doi":"10.21203/rs.3.rs-8872718/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"101014778361650530771886115899349322158","date":"2026-03-31T15:59:07+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"258420339988445764739523369755762091394","date":"2026-02-19T14:47:31+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-19T11:09:02+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-17T07:38:22+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-17T07:36:22+00:00","index":"","fulltext":""},{"type":"submitted","content":"Neural Processing Letters","date":"2026-02-13T13:57:08+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"neural-processing-letters","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"nepl","sideBox":"Learn more about [Neural Processing Letters](http://link.springer.com/journal/11063)","snPcode":"11063","submissionUrl":"https://submission.nature.com/new-submission/11063/3","title":"Neural Processing Letters","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"6de34e4c-17c5-4c4b-9dd8-b1fd4a430dd3","owner":[],"postedDate":"February 24th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-02-24T15:00:06+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-24 15:00:06","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8872718","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8872718","identity":"rs-8872718","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}