{"data":{"id":"10.5281/zenodo.23752393","type":"dois","attributes":{"doi":"10.5281/zenodo.23752393","prefix":"10.5281","suffix":"zenodo.23752393","identifiers":[{"identifier":"oai:zenodo.org:23752393","identifierType":"oai"}],"alternateIdentifiers":[{"alternateIdentifierType":"oai","alternateIdentifier":"oai:zenodo.org:23752393"}],"creators":[{"name":"Armani, Ryan","nameType":"Personal","givenName":"Ryan","familyName":"Armani","affiliation":["Independent researcher"],"nameIdentifiers":[{"nameIdentifier":"0009-0005-9964-6841","nameIdentifierScheme":"ORCID"}]}],"titles":[{"title":"Cost per Verified Task: Replay-Based Evaluation of Back-Office Agents"}],"publisher":"Zenodo","container":{},"publicationYear":2026,"subjects":[{"subject":"agent evaluation"},{"subject":"cost per task"},{"subject":"deterministic replay"},{"subject":"simulated population"},{"subject":"back-office tasks"}],"contributors":[],"dates":[{"date":"2026-10-29","dateType":"Issued"}],"language":null,"types":{"ris":"GEN","bibtex":"misc","citeproc":"article","schemaOrg":"CreativeWork","resourceType":"","resourceTypeGeneral":"Preprint"},"relatedIdentifiers":[{"relationType":"IsSupplementedBy","relatedIdentifier":"https://github.com/pooriaarab/benchmark-cost-per-task","resourceTypeGeneral":"Software","relatedIdentifierType":"URL"},{"relationType":"IsVersionOf","relatedIdentifier":"10.5281/zenodo.23752392","relatedIdentifierType":"DOI"}],"relatedItems":[],"sizes":[],"formats":[],"version":null,"rightsList":[{"rights":"Creative Commons Attribution 4.0 International","rightsUri":"https://creativecommons.org/licenses/by/4.0/legalcode","schemeUri":"https://spdx.org/licenses/","rightsIdentifier":"cc-by-4.0","rightsIdentifierScheme":"SPDX"}],"descriptions":[{"description":"Cost per attempted task can conceal expensive failures. We study cost per verified success on 30 synthetic back-office tasks, using deterministic replay of file and customer-record operations. A privileged oracle completes every task (95% Wilson interval: 88.6-100.0%); each of six trivial policies completes none (0.0-11.4%). Ninety invalid-trace probes fail verification. A simulated population sweep compares direct execution with a redundant read-only control under a fixed artificial token tariff. The control preserves success and doubles assumed cost per success. These results validate accounting and replay behavior, not learned-agent ability or provider prices. No hosted model or human study is included.","descriptionType":"Abstract"}],"geoLocations":[],"fundingReferences":[],"xml":"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","url":"https://zenodo.org/doi/10.5281/zenodo.23752393","contentUrl":null,"metadataVersion":2,"schemaVersion":"http://datacite.org/schema/kernel-4","source":"api","isActive":true,"state":"findable","reason":null,"viewCount":0,"viewsOverTime":[],"downloadCount":0,"downloadsOverTime":[],"referenceCount":0,"citationCount":0,"citationsOverTime":[],"partCount":0,"partOfCount":0,"versionCount":0,"versionOfCount":1,"created":"2026-10-29T03:17:11.000Z","registered":"2026-10-29T03:17:12.000Z","published":"2026","updated":"2026-11-07T05:37:14.000Z"},"relationships":{"client":{"data":{"id":"cern.zenodo","type":"clients"}},"provider":{"data":{"id":"cern","type":"providers"}},"media":{"data":{"id":"10.5281/zenodo.23752393","type":"media"}},"references":{"data":[]},"citations":{"data":[]},"parts":{"data":[]},"partOf":{"data":[]},"versions":{"data":[]},"versionOf":{"data":[{"id":"10.5281/zenodo.23752392","type":"dois"}]}}}}