Evaluating Abstract Reasoning and Problem-Solving Abilities of Large Language Models Using Raven's Progressive Matrices

Evaluating Abstract Reasoning and Problem-Solving Abilities of Large Language Models Using Raven's Progressive Matrices | 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 Evaluating Abstract Reasoning and Problem-Solving Abilities of Large Language Models Using Raven's Progressive Matrices Chengru Zhang, Liuyun Wang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4560345/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Artificial intelligence has rapidly evolved, leading to the development of powerful models capable of performing complex cognitive tasks. Evaluating the cognitive abilities of these models through established human intelligence tests such as Raven's Progressive Matrices (RPM) offers a novel and significant approach to understanding their abstract reasoning capabilities. The study adapted RPM for text-based interactions, enabling the evaluation of Mistral and Llama without human intervention. Results revealed that both models surpass average human performance in overall accuracy, demonstrating their advanced problem-solving skills. However, the analysis also highlighted variability in performance across different types of reasoning tasks, with Llama excelling in sequential pattern recognition and Mistral showing weaknesses in spatial awareness. These findings provide valuable insights into the strengths and limitations of Mistral and Llama, offering a comprehensive understanding of their cognitive abilities and guiding future advancements in artificial intelligence. Artificial Intelligence and Machine Learning Cognitive evaluation Abstract reasoning Problem-solving Artificial intelligence RPM test Language models Full Text Additional Declarations The authors declare no competing interests. Cite Share Download PDF Status: Posted Version 1 posted 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-4560345","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":312862594,"identity":"bc9cf9eb-934c-4074-a797-e094b4616e2d","order_by":0,"name":"Chengru Zhang","email":"data:image/png;base64,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","orcid":"https://orcid.org/0009-0000-0334-5440","institution":"","correspondingAuthor":true,"prefix":"","firstName":"Chengru","middleName":"","lastName":"Zhang","suffix":""},{"id":312862595,"identity":"bf5ec845-483a-487d-8eab-6143d33a0ea7","order_by":1,"name":"Liuyun Wang","email":"","orcid":"","institution":"","correspondingAuthor":false,"prefix":"","firstName":"Liuyun","middleName":"","lastName":"Wang","suffix":""}],"badges":[],"createdAt":"2024-06-10 23:51:49","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-4560345/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4560345/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":58118251,"identity":"56d121d0-e7db-43bc-8aaa-7273488ca3b9","added_by":"auto","created_at":"2024-06-11 11:25:06","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":187307,"visible":true,"origin":"","legend":"","description":"","filename":"evaluating.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4560345/v1_covered_785e6818-5e15-42b1-9b0f-0b2fa07b55ce.pdf"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003eEvaluating Abstract Reasoning and Problem-Solving Abilities of Large Language Models Using Raven's Progressive Matrices\u003c/p\u003e","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"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":"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":"Cognitive evaluation, Abstract reasoning, Problem-solving, Artificial intelligence, RPM test, Language models","lastPublishedDoi":"10.21203/rs.3.rs-4560345/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4560345/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eArtificial intelligence has rapidly evolved, leading to the development of powerful models capable of performing complex cognitive tasks. Evaluating the cognitive abilities of these models through established human intelligence tests such as Raven's Progressive Matrices (RPM) offers a novel and significant approach to understanding their abstract reasoning capabilities. The study adapted RPM for text-based interactions, enabling the evaluation of Mistral and Llama without human intervention. Results revealed that both models surpass average human performance in overall accuracy, demonstrating their advanced problem-solving skills. However, the analysis also highlighted variability in performance across different types of reasoning tasks, with Llama excelling in sequential pattern recognition and Mistral showing weaknesses in spatial awareness. These findings provide valuable insights into the strengths and limitations of Mistral and Llama, offering a comprehensive understanding of their cognitive abilities and guiding future advancements in artificial intelligence.\u003c/p\u003e","manuscriptTitle":"Evaluating Abstract Reasoning and Problem-Solving Abilities of Large Language Models Using Raven's Progressive Matrices","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-06-11 11:16:59","doi":"10.21203/rs.3.rs-4560345/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":"7b4650df-51ec-431a-82b2-c2604543c6e1","owner":[],"postedDate":"June 11th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":33073949,"name":"Artificial Intelligence and Machine Learning"}],"tags":[],"updatedAt":"2024-06-11T11:16:59+00:00","versionOfRecord":[],"versionCreatedAt":"2024-06-11 11:16:59","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4560345","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4560345","identity":"rs-4560345","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}