The Art of Repair in Human-Agent Conversations: A Taxonomy of Repair Strategies by Users and LLM-Based Conversational Agents

The Art of Repair in Human-Agent Conversations: A Taxonomy of Repair Strategies by Users and LLM-Based Conversational Agents | 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 The Art of Repair in Human-Agent Conversations: A Taxonomy of Repair Strategies by Users and LLM-Based Conversational Agents Gunnar Stevens, Delong Korus-Du, Alexander Boden, Peter Tolmie, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7301450/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 12 You are reading this latest preprint version Abstract Large Language Models (LLMs) like ChatGPT are increasingly embedded in everyday tasks of various professions, yet LLMs' outputs often remain unreliable, ambiguous, or misleading. This paper explores how users identify and repair troubles to make LLMs make things right in situated interactions of various contexts. Using an ethnomethodological lens, we examine 21 real-world chat transcripts across diverse work contexts. Our analysis reveals a broad repertoire of repair practices, including factual corrections, stylistic refinements, implicit signals, and strategic reframings. The findings challenge the view that users’ repair work on LLM outputs is merely a response to system failure. Instead, our findings present a taxonomy of repair work of both users and conversational agents, comprising 6 types of repair initiators (errors, dissatisfactions, apologies, shortcomings, implicit signals, and contextualization), 3 stages of repair elements (6 types of trouble classification, 3 types of trouble specification, and 7 types of trouble management), and 3 types of repair processes (incremental, grounding, and validating). These repair categories demonstrate the core of human-agent collaboration: meaning and correctness are not pre-given but are achieved through situated work for all practical purposes. By treating trouble as an ordinary part of collaborative work, we highlight the need to design for user repair interaction alongside improving model reliability. These findings contribute to ongoing debates in HCI and CSCW around the accountability, intelligibility, and co-construction of meaning in human-AI interaction regarding LLMs applications. Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 24 Feb, 2026 Reviews received at journal 13 Feb, 2026 Reviewers agreed at journal 07 Feb, 2026 Reviews received at journal 03 Feb, 2026 Reviewers agreed at journal 07 Jan, 2026 Reviews received at journal 19 Dec, 2025 Reviewers agreed at journal 05 Oct, 2025 Reviewers agreed at journal 12 Sep, 2025 Reviewers invited by journal 25 Aug, 2025 Editor assigned by journal 19 Aug, 2025 Submission checks completed at journal 11 Aug, 2025 First submitted to journal 05 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-7301450","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":508349434,"identity":"88d2404c-38a3-4f53-9351-63793cb4e531","order_by":0,"name":"Gunnar Stevens","email":"","orcid":"","institution":"University of Siegen","correspondingAuthor":false,"prefix":"","firstName":"Gunnar","middleName":"","lastName":"Stevens","suffix":""},{"id":508349435,"identity":"b99cf18c-a3f1-4d2e-9c8c-4c2d9d4ef9b1","order_by":1,"name":"Delong Korus-Du","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABBUlEQVRIiWNgGAWjYDCCA8wNIIqxgR1IJzAw8ECE2fBpYYRq4TlAshaJBGRhPFr4jjc2Pi74xSC74eYbswcPGOpk+PkXH3vAUGaDU4vkmYPNxjP7GIw33M4xN0hgOMwjOeNZugHDuTScWgxuJLZJ8/YwJAK1mEkk/jvAY3DjjJkEY9th3FruP4RquQlUmcBQx2N/4/w3oJb/eGxhbJPm+QHUcoMHpIWZx4C/hw2o5QAevyQ2G/M2SBjPPJNWJgHyi8QNNqDec8k4tfAdP3zwMc8fG9m+44e3Sf5gqLPn7z/8TOJDmR1OLWDA2CaBxANFUAJ+DUDwB5nDj9sbo2AUjIJRMDIBAL2xVrMVYblhAAAAAElFTkSuQmCC","orcid":"","institution":"University of Siegen","correspondingAuthor":true,"prefix":"","firstName":"Delong","middleName":"","lastName":"Korus-Du","suffix":""},{"id":508349436,"identity":"18173617-8b0b-42b4-b45a-c68efd0e6ad6","order_by":2,"name":"Alexander Boden","email":"","orcid":"","institution":"Hochschule Bonn-Rhein-Sieg University of Applied Sciences","correspondingAuthor":false,"prefix":"","firstName":"Alexander","middleName":"","lastName":"Boden","suffix":""},{"id":508349437,"identity":"f910411c-eba6-4949-ab15-945cf588216b","order_by":3,"name":"Peter Tolmie","email":"","orcid":"","institution":"University of Siegen","correspondingAuthor":false,"prefix":"","firstName":"Peter","middleName":"","lastName":"Tolmie","suffix":""},{"id":508349438,"identity":"fb635f73-223c-46d5-bda4-76c752a9da11","order_by":4,"name":"Dave Randall","email":"","orcid":"","institution":"University of Siegen","correspondingAuthor":false,"prefix":"","firstName":"Dave","middleName":"","lastName":"Randall","suffix":""},{"id":508349439,"identity":"df641f24-ebdd-437f-9245-8b21980ac772","order_by":5,"name":"Md Shajalal","email":"","orcid":"","institution":"University of Siegen","correspondingAuthor":false,"prefix":"","firstName":"Md","middleName":"","lastName":"Shajalal","suffix":""}],"badges":[],"createdAt":"2025-08-05 14:08:08","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7301450/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7301450/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":90377817,"identity":"c353c08e-5e5e-48f8-8750-8348f377bd76","added_by":"auto","created_at":"2025-09-02 06:17:12","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1290657,"visible":true,"origin":"","legend":"","description":"","filename":"SpringerJCSCWTheArtofRepairinHumanAgentConversationsATaxonomyofRepairStrategiesbyUsersandLLMBasedConversationalAgents3.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7301450/v1_covered_2357c9ce-6b97-42b0-b4e9-7c2f0276568a.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"The Art of Repair in Human-Agent Conversations: A Taxonomy of Repair Strategies by Users and LLM-Based Conversational Agents","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":"computer-supported-cooperative-work-cscw","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"cosu","sideBox":"Learn more about [Computer Supported Cooperative Work (CSCW)](http://link.springer.com/journal/10606)","snPcode":"10606","submissionUrl":"https://submission.nature.com/new-submission/10606/3","title":"Computer Supported Cooperative Work (CSCW)","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7301450/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7301450/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eLarge Language Models (LLMs) like ChatGPT are increasingly embedded in everyday tasks of various professions, yet LLMs' outputs often remain unreliable, ambiguous, or misleading. This paper explores how users identify and repair troubles to make LLMs make things right in situated interactions of various contexts. Using an ethnomethodological lens, we examine 21 real-world chat transcripts across diverse work contexts. Our analysis reveals a broad repertoire of repair practices, including factual corrections, stylistic refinements, implicit signals, and strategic reframings. The findings challenge the view that users’ repair work on LLM outputs is merely a response to system failure. Instead, our findings present a taxonomy of repair work of both users and conversational agents, comprising \u0026nbsp;6 types of repair initiators (errors, dissatisfactions, apologies, shortcomings, implicit signals, and contextualization), 3 stages of repair elements (6 types of trouble classification, 3 types of trouble specification, and 7 types of trouble management), and 3 types of repair processes (incremental, grounding, and validating). These repair categories demonstrate the core of human-agent collaboration: meaning and correctness are not pre-given but are achieved through situated work for all practical purposes. By treating trouble as an ordinary part of collaborative work, we highlight the need to design for user repair interaction alongside improving model reliability. These findings contribute to ongoing debates in HCI and CSCW around the accountability, intelligibility, and co-construction of meaning in human-AI interaction regarding LLMs applications.\u003c/p\u003e","manuscriptTitle":"The Art of Repair in Human-Agent Conversations: A Taxonomy of Repair Strategies by Users and LLM-Based Conversational Agents","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-02 06:09:06","doi":"10.21203/rs.3.rs-7301450/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-02-24T08:17:29+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-13T17:26:51+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"192150058267291623570411642799505076153","date":"2026-02-07T14:57:25+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-02-03T14:44:26+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"221796057061311940597764138220314993587","date":"2026-01-07T12:52:33+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-19T23:10:41+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"115206111614853799281159624082429262096","date":"2025-10-05T14:42:02+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"83837425758082008434789881505147698466","date":"2025-09-12T08:10:24+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-08-25T08:11:24+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-08-19T14:51:56+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-08-11T04:42:45+00:00","index":"","fulltext":""},{"type":"submitted","content":"Computer Supported Cooperative Work (CSCW)","date":"2025-08-05T13:52:24+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"computer-supported-cooperative-work-cscw","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"cosu","sideBox":"Learn more about [Computer Supported Cooperative Work (CSCW)](http://link.springer.com/journal/10606)","snPcode":"10606","submissionUrl":"https://submission.nature.com/new-submission/10606/3","title":"Computer Supported Cooperative Work (CSCW)","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"56e64c8b-4ae9-4711-b585-cd82ca5641ad","owner":[],"postedDate":"September 2nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[],"tags":[],"updatedAt":"2026-02-24T08:24:24+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-02 06:09:06","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7301450","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7301450","identity":"rs-7301450","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}