{"status":"OK","data":{"id":648045,"identifier":"DAHVMQ","persistentUrl":"https://doi.org/10.34894/DAHVMQ","protocol":"doi","authority":"10.34894","separator":"/","publisher":"DataverseNL","publicationDate":"2026-08-31","storageIdentifier":"surf://10.34894/DAHVMQ","effectiveDatasetFileCountLimit":10000,"datasetFileUploadsAvailable":9997,"datasetType":"dataset","locks":[],"latestVersion":{"id":35284,"datasetId":648045,"datasetPersistentId":"doi:10.34894/DAHVMQ","datasetType":"dataset","storageIdentifier":"surf://10.34894/DAHVMQ","versionNumber":2,"internalVersionNumber":26,"versionMinorNumber":0,"versionState":"RELEASED","latestVersionPublishingState":"RELEASED","productionDate":"2026-06-14","lastUpdateTime":"2026-09-02T15:24:05Z","releaseTime":"2026-09-02T15:24:05Z","createTime":"2026-08-31T12:43:37Z","publicationDate":"2026-08-31","citationDate":"2026-08-31","effectiveDatasetFileCountLimit":10000,"datasetFileUploadsAvailable":9997,"license":{"name":"CC-BY-4.0","uri":"http://creativecommons.org/licenses/by/4.0","iconUri":"https://i.creativecommons.org/l/by/4.0/88x31.png"},"fileAccessRequest":true,"metadataBlocks":{"citation":{"displayName":"Citation Metadata","name":"citation","fields":[{"typeName":"title","multiple":false,"typeClass":"primitive","value":"DiagFlowBench"},{"typeName":"subtitle","multiple":false,"typeClass":"primitive","value":"Evaluating How Language Models Handle Off-Procedure Inputs in Grounded Diagnostic Dialogue"},{"typeName":"author","multiple":true,"typeClass":"compound","value":[{"authorName":{"typeName":"authorName","multiple":false,"typeClass":"primitive","value":"Gil de Avalle, Guillermo"},"authorAffiliation":{"typeName":"authorAffiliation","multiple":false,"typeClass":"primitive","value":"https://ror.org/012p63287","expandedvalue":{"scheme":"http://www.grid.ac/ontology/","termName":"University of Groningen","@type":"https://schema.org/Organization"}},"authorIdentifierScheme":{"typeName":"authorIdentifierScheme","multiple":false,"typeClass":"controlledVocabulary","value":"ORCID"},"authorIdentifier":{"typeName":"authorIdentifier","multiple":false,"typeClass":"primitive","value":"0009-0004-1538-2256"}},{"authorName":{"typeName":"authorName","multiple":false,"typeClass":"primitive","value":"Maruster, Laura"},"authorAffiliation":{"typeName":"authorAffiliation","multiple":false,"typeClass":"primitive","value":"https://ror.org/012p63287","expandedvalue":{"scheme":"http://www.grid.ac/ontology/","termName":"University of Groningen","@type":"https://schema.org/Organization"}},"authorIdentifierScheme":{"typeName":"authorIdentifierScheme","multiple":false,"typeClass":"controlledVocabulary","value":"ORCID"},"authorIdentifier":{"typeName":"authorIdentifier","multiple":false,"typeClass":"primitive","value":"0000-0002-6588-7648"}},{"authorName":{"typeName":"authorName","multiple":false,"typeClass":"primitive","value":"Emmanouilidis, Christos"},"authorAffiliation":{"typeName":"authorAffiliation","multiple":false,"typeClass":"primitive","value":"https://ror.org/012p63287","expandedvalue":{"scheme":"http://www.grid.ac/ontology/","termName":"University of Groningen","@type":"https://schema.org/Organization"}},"authorIdentifierScheme":{"typeName":"authorIdentifierScheme","multiple":false,"typeClass":"controlledVocabulary","value":"ORCID"},"authorIdentifier":{"typeName":"authorIdentifier","multiple":false,"typeClass":"primitive","value":"0000-0003-4335-6915"}}]},{"typeName":"datasetContact","multiple":true,"typeClass":"compound","value":[{"datasetContactName":{"typeName":"datasetContactName","multiple":false,"typeClass":"primitive","value":"Emmanouilidis, Christos"},"datasetContactAffiliation":{"typeName":"datasetContactAffiliation","multiple":false,"typeClass":"primitive","value":"rug.nl"}},{"datasetContactName":{"typeName":"datasetContactName","multiple":false,"typeClass":"primitive","value":"Groningen Digital Competence Centre"},"datasetContactAffiliation":{"typeName":"datasetContactAffiliation","multiple":false,"typeClass":"primitive","value":"University of Groningen"}}]},{"typeName":"dsDescription","multiple":true,"typeClass":"compound","value":[{"dsDescriptionValue":{"typeName":"dsDescriptionValue","multiple":false,"typeClass":"primitive","value":"DiagFlowBench is a benchmark for evaluating how language models handle off-procedure operator inputs in flowchart-grounded diagnostic dialogue. It comprises 50 anonymised troubleshooting and operator-control flowcharts drawn from the maintenance documentation of a consumer goods manufacturer, together with 1,676 multi-turn conversations rendered from them. Fault classes include conveyor belt tracking, vision inspection calibration, robotic placement, and CNC spindle acceptance. All proprietary identifiers have been removed and the domain abstracted to a generic industrial setting, preserving the diagnostic logic and graph topology of the source procedures.<br><br>\n\nEach conversation is a turn-by-turn exchange between a factory operator and an advisory system working through one of the flowcharts. In a cooperative conversation the operator reports only things the flowchart anticipates. At every turn there is therefore a documented step the system can point to next. In an off-procedure conversation the operator at some point reports something the flowchart has no branch for, such as a symptom that was never documented, a fault in a component the procedure does not mention, or a question about an unrelated matter. The system should then recognise that the procedure does not cover the situation rather than pressing on with the next documented step. The two kinds are released as 838 matched pairs (1,676 conversations in total). Within a pair, the cooperative conversation and its counterpart differ only in whether the operator leaves the procedure at some point.<br><br>\n\nThe dataset accompanies the paper DiagFlowBench: Evaluating How Language Models Handle Off-Procedure Inputs in Grounded Diagnostic Dialogue, accepted to the Industry Track of the 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP 2026).<br><br>\n\nGeneration and evaluation code, prompts and pinned model identifiers are not included in this deposit and are released under MIT at https://github.com/guille-gil/DiagFlowBench.\n\nNote that although the repository unpacks files compressed within the dataset.zip file, there is no need to check for unpacked files in the repository: you only need to download the dataset.zip compressed file, which contains everything else."},"dsDescriptionDate":{"typeName":"dsDescriptionDate","multiple":false,"typeClass":"primitive","value":"2026-06-14"}}]},{"typeName":"subject","multiple":true,"typeClass":"controlledVocabulary","value":["Business and Management","Computer and Information Science","Engineering"]},{"typeName":"keyword","multiple":true,"typeClass":"compound","value":[{"keywordValue":{"typeName":"keywordValue","multiple":false,"typeClass":"primitive","value":"Benchmarking data"}},{"keywordValue":{"typeName":"keywordValue","multiple":false,"typeClass":"primitive","value":"Language model"}},{"keywordValue":{"typeName":"keywordValue","multiple":false,"typeClass":"primitive","value":"Diagnostics"}},{"keywordValue":{"typeName":"keywordValue","multiple":false,"typeClass":"primitive","value":"Dialogue systems"}}]},{"typeName":"publication","multiple":true,"typeClass":"compound","value":[{"publicationRelationType":{"typeName":"publicationRelationType","multiple":false,"typeClass":"controlledVocabulary","value":"IsSupplementTo"},"publicationCitation":{"typeName":"publicationCitation","multiple":false,"typeClass":"primitive","value":"DiagFlowBench: Evaluating How Language Models Handle Off-Procedure Inputs in Grounded Diagnostic Dialogue, accepted to the Industry Track of the 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP 2026)"}}]},{"typeName":"language","multiple":true,"typeClass":"controlledVocabulary","value":["English"]},{"typeName":"productionDate","multiple":false,"typeClass":"primitive","value":"2026-06-14"},{"typeName":"contributor","multiple":true,"typeClass":"compound","value":[{"contributorType":{"typeName":"contributorType","multiple":false,"typeClass":"controlledVocabulary","value":"Project Leader"},"contributorName":{"typeName":"contributorName","multiple":false,"typeClass":"primitive","value":"1) Emmanouilidis, Christos; 2) University of Groningen"}},{"contributorType":{"typeName":"contributorType","multiple":false,"typeClass":"controlledVocabulary","value":"Data Curator"},"contributorName":{"typeName":"contributorName","multiple":false,"typeClass":"primitive","value":"1) de Avalle, G.G.; 2) University of Groningen"}}]},{"typeName":"grantNumber","multiple":true,"typeClass":"compound","value":[{"grantNumberAgency":{"typeName":"grantNumberAgency","multiple":false,"typeClass":"primitive","value":"Horizon Europe"},"grantNumberValue":{"typeName":"grantNumberValue","multiple":false,"typeClass":"primitive","value":"101214389"}}]},{"typeName":"distributor","multiple":true,"typeClass":"compound","value":[{"distributorName":{"typeName":"distributorName","multiple":false,"typeClass":"primitive","value":"DataverseNL Network"},"distributorAffiliation":{"typeName":"distributorAffiliation","multiple":false,"typeClass":"primitive","value":"DANS"},"distributorAbbreviation":{"typeName":"distributorAbbreviation","multiple":false,"typeClass":"primitive","value":"DVNL"},"distributorURL":{"typeName":"distributorURL","multiple":false,"typeClass":"primitive","value":"https://dataverse.nl"}}]},{"typeName":"depositor","multiple":false,"typeClass":"primitive","value":"Emmanouilidis, Christos"},{"typeName":"dateOfDeposit","multiple":false,"typeClass":"primitive","value":"2026-06-14"},{"typeName":"software","multiple":true,"typeClass":"compound","value":[{"softwareName":{"typeName":"softwareName","multiple":false,"typeClass":"primitive","value":"DiagFlowBench"}}]},{"typeName":"relatedMaterial","multiple":true,"typeClass":"primitive","value":["Generation and evaluation code, prompts and pinned model identifiers are not included in this deposit and are released under MIT license at <a href=\"https://github.com/guille-gil/DiagFlowBench\">https://github.com/guille-gil/DiagFlowBench</a>"]}]},"dansDataVaultMetadata":{"displayName":"Data Vault Metadata","name":"dansDataVaultMetadata","fields":[{"typeName":"dansDataversePid","multiple":false,"typeClass":"primitive","value":"doi:10.34894/DAHVMQ"},{"typeName":"dansDataversePidVersion","multiple":false,"typeClass":"primitive","value":"2.0"},{"typeName":"dansBagId","multiple":false,"typeClass":"primitive","value":"urn:uuid:55d13c0d-5cc3-4a68-a50e-5d92083e5cd1"},{"typeName":"dansNbn","multiple":false,"typeClass":"primitive","value":"urn:nbn:nl:ui:13-e9c5a5f1-be24-42f4-8e05-8f076bc8c2cf"}]}},"files":[{"description":"The complete data set in a compressed file","label":"data.zip","restricted":false,"directoryLabel":"dataset","version":1,"datasetVersionId":35284,"dataFile":{"id":672957,"persistentId":"","filename":"data.zip","contentType":"application/zip","friendlyType":"ZIP Archive","filesize":44319066,"description":"The complete data set in a compressed file","storageIdentifier":"surf://store:1a05c210e9e-dfd1cfc2a128","rootDataFileId":-1,"checksum":{"type":"SHA-1","value":"f274c8640d9d7b130afdf0ce632397cf02526b96"},"tabularData":false,"creationDate":"2026-09-01","publicationDate":"2026-09-02","directoryLabel":"dataset","lastUpdateTime":"2026-09-02T15:24:05Z","fileAccessRequest":true}},{"description":"Readme file describing the details and usage of the dataset","label":"README.md","restricted":false,"directoryLabel":"dataset","version":1,"datasetVersionId":35284,"dataFile":{"id":672958,"persistentId":"","filename":"README.md","contentType":"text/markdown","friendlyType":"Markdown Text","filesize":4523,"description":"Readme file describing the details and usage of the dataset","storageIdentifier":"surf://store:1a05c210eea-8ddec80d9a7c","rootDataFileId":-1,"checksum":{"type":"SHA-1","value":"f471d181e5ad962b17c8e0aee6387b05f288f5d3"},"tabularData":false,"creationDate":"2026-09-01","publicationDate":"2026-09-02","directoryLabel":"dataset","lastUpdateTime":"2026-09-02T15:24:05Z","fileAccessRequest":true}}]}}}