ARAFA: An LLM Generated Arabic Fact-Checking Dataset

ARAFA: An LLM Generated Arabic Fact-Checking Dataset | 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 ARAFA: An LLM Generated Arabic Fact-Checking Dataset Christophe Khalil, Shady Elbassuoni, Rida Assaf This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7335564/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 12 You are reading this latest preprint version Abstract Automatic fact-checking poses a significant challenge in Arabic natural language processing due to the scarcity of datasets and resources. In this manuscript, we introduce ARAFA, a new large-scale dataset for fact-checking in Modern Standard Arabic, constructed through an automated framework leveraging large language models (LLMs). The dataset was constructed through a three-step pipeline: (1)claim generation from Arabic Wikipedia pages with supporting textual evidence,(2) claim mutation to generate challenging counterfactual claims with refuting ev-idence, and (3) an automatic validation step to validate that the generated claimsare either supported or refuted by their accompanying evidence, or if the evidencedoes not provide enough information to judge the validity of the claims. The resulting dataset comprises 181,976 claim-evidence pairs labeled as supported, refuted, or not enough information. Human evaluation carried out on a test sample from the dataset demonstrated strong inter-annotator agreement (κ = 0.89)using Cohen’s Kappa for supported claims and (κ = 0.94) for refuted claims. Automatic validation based on human-evaluated sample achieved 86% accuracy for supported claims and 88% for refuted ones. To showcase ARAFA’s value as a resource for automatic Arabic fact-checking, four open-source transformer-based models were fine-tuned using ARAFA, with the top-performing model achieving a Macro F1-score of 77% on the test data. In addition to ARAFA being the first large-scale dataset for Arabic fact-checking, our framework presents a scalable approach for developing similar resources for other low-resource languages. fact-checking Arabic NLP claim verification evidence retrieval Full Text Additional Declarations No competing interests reported. Supplementary Files ARAFASupplementaryMaterial.pdf Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 12 Dec, 2025 Reviews received at journal 25 Sep, 2025 Reviews received at journal 19 Sep, 2025 Reviews received at journal 18 Sep, 2025 Reviewers agreed at journal 25 Aug, 2025 Reviewers agreed at journal 23 Aug, 2025 Reviewers agreed at journal 23 Aug, 2025 Reviewers agreed at journal 22 Aug, 2025 Reviewers invited by journal 22 Aug, 2025 Editor assigned by journal 22 Aug, 2025 Submission checks completed at journal 11 Aug, 2025 First submitted to journal 09 Aug, 2025 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-7335564","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":498145167,"identity":"d2203779-bb33-4e0b-bb40-dee832ef0b10","order_by":0,"name":"Christophe Khalil","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA00lEQVRIiWNgGAWjYFACNsYHHyqAFDuQzdgAJA4Q1sJsOOMMiCJBC5s0bxuQJloLf/uxBGneedvk+ZgZGD/z7mCQ47uRgF+LxJm0A4Zzt902bGNmYJbmPcNgLElIiwFDekPC2223GYFaGEAuTNxAUAv/84YDvHNu24Ns+Q3UUk9Yi0TawUbehtuJQC3gcEgwIOiXG8+SGWccu53cxszYZjn3jIThzDMP8Gvh708z//Gh5rbt/Pbmwzfe7rCR5ztOwBYkAI4UCaKVj4JRMApGwSjAAwDWUUNkLxrNqQAAAABJRU5ErkJggg==","orcid":"","institution":"American University of Beirut","correspondingAuthor":true,"prefix":"","firstName":"Christophe","middleName":"","lastName":"Khalil","suffix":""},{"id":498145168,"identity":"84e84975-f3b8-4934-95d7-cfd8a0db00a6","order_by":1,"name":"Shady Elbassuoni","email":"","orcid":"","institution":"American University of Beirut","correspondingAuthor":false,"prefix":"","firstName":"Shady","middleName":"","lastName":"Elbassuoni","suffix":""},{"id":498145169,"identity":"ac12ab52-7d3e-4975-bf86-9c9021897712","order_by":2,"name":"Rida Assaf","email":"","orcid":"","institution":"American University of Beirut","correspondingAuthor":false,"prefix":"","firstName":"Rida","middleName":"","lastName":"Assaf","suffix":""}],"badges":[],"createdAt":"2025-08-09 18:53:08","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7335564/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7335564/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":88875641,"identity":"482058ce-0a36-4a82-88d1-c6107d6b14cf","added_by":"auto","created_at":"2025-08-12 10:05:19","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1479543,"visible":true,"origin":"","legend":"","description":"","filename":"ARAFA.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7335564/v1_covered_769600f9-650a-4095-9ca2-31969e3ca1ac.pdf"},{"id":88875322,"identity":"28f5a55b-e564-43cd-8e4d-186c22b32821","added_by":"auto","created_at":"2025-08-12 09:57:16","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":190705,"visible":true,"origin":"","legend":"","description":"","filename":"ARAFASupplementaryMaterial.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7335564/v1/dd4fa5855a419c25c6790949.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"ARAFA: An LLM Generated Arabic Fact-Checking Dataset","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"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":"language-resources-and-evaluation","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"lrev","sideBox":"Learn more about [Language Resources and Evaluation](http://link.springer.com/journal/10579)","snPcode":"10579","submissionUrl":"https://submission.nature.com/new-submission/10579/3","title":"Language Resources and Evaluation","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"fact-checking, Arabic NLP, claim verification, evidence retrieval","lastPublishedDoi":"10.21203/rs.3.rs-7335564/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7335564/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Automatic fact-checking poses a significant challenge in Arabic natural language processing due to the scarcity of datasets and resources. In this manuscript, we introduce ARAFA, a new large-scale dataset for fact-checking in Modern Standard Arabic, constructed through an automated framework leveraging large language models (LLMs). The dataset was constructed through a three-step pipeline: (1)claim generation from Arabic Wikipedia pages with supporting textual evidence,(2) claim mutation to generate challenging counterfactual claims with refuting ev-idence, and (3) an automatic validation step to validate that the generated claimsare either supported or refuted by their accompanying evidence, or if the evidencedoes not provide enough information to judge the validity of the claims. The resulting dataset comprises 181,976 claim-evidence pairs labeled as supported, refuted, or not enough information. Human evaluation carried out on a test sample from the dataset demonstrated strong inter-annotator agreement (κ = 0.89)using Cohen’s Kappa for supported claims and (κ = 0.94) for refuted claims. Automatic validation based on human-evaluated sample achieved 86% accuracy for supported claims and 88% for refuted ones. To showcase ARAFA’s value as a resource for automatic Arabic fact-checking, four open-source transformer-based models were fine-tuned using ARAFA, with the top-performing model achieving a Macro F1-score of 77% on the test data. In addition to ARAFA being the first large-scale dataset for Arabic fact-checking, our framework presents a scalable approach for developing similar resources for other low-resource languages.","manuscriptTitle":"ARAFA: An LLM Generated Arabic Fact-Checking Dataset","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-12 09:57:11","doi":"10.21203/rs.3.rs-7335564/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-12-12T08:25:09+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-25T19:25:11+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-20T01:37:50+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-09-18T20:00:11+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"109993441143049936021697012461946169112","date":"2025-08-25T13:44:59+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"214535699247090226839462473023255686853","date":"2025-08-23T10:17:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"210621683260304816332243810389653595900","date":"2025-08-23T09:23:50+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"7722190900046960045121883740169309467","date":"2025-08-22T19:10:39+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-22T13:56:53+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-22T13:07:04+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-12T02:57:09+00:00","index":"","fulltext":""},{"type":"submitted","content":"Language Resources and Evaluation","date":"2025-08-09T18:40:46+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"language-resources-and-evaluation","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"lrev","sideBox":"Learn more about [Language Resources and Evaluation](http://link.springer.com/journal/10579)","snPcode":"10579","submissionUrl":"https://submission.nature.com/new-submission/10579/3","title":"Language Resources and Evaluation","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"14d52d9a-053c-455c-b9fe-e67c9c88af84","owner":[],"postedDate":"August 12th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-01-19T17:52:55+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-12 09:57:11","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7335564","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7335564","identity":"rs-7335564","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}