{"@context":"https://schema.org","@type":"Dataset","name":"Science AI Journal Calibration Corpus","url":"https://scienceaijournal.com/dataset","license":"https://creativecommons.org/licenses/by/4.0/","licenseScope":"The CC BY 4.0 licence covers this aggregate summary data only. The peer-review texts it counts were published by their source platforms and remain under those platforms' terms.","inLanguage":"en","generatedAt":"2026-11-18T19:42:04.033Z","counts":{"reviews":5000,"reviewsUnit":"distinct peer-review texts (display floor)","sourcePlatforms":15,"agents":9,"libraryPapers":7900000,"openAlexWorks":250000000,"indexedGaps":1100000,"gapPages":1900},"measured":{"at":"2026-10-28","unit":"one distinct peer-review text: reviewText normalised (whitespace collapsed, lower-cased, >= 50 chars) and hashed, minus synthetic sample_ texts, minus fragments contained in a longer text of the same paper, times the hand-classified genuine share (referee reports and fields, reviewer follow-ups, editor decisions and assessments, public comments, author responses)","estimate":5598,"conservativeFloor":5092,"reviewerWrittenEstimate":4184,"perPlatform":[{"platform":"OpenReview","texts":3033},{"platform":"SciPost","texts":818},{"platform":"PREreview","texts":473},{"platform":"PeerJ","texts":302},{"platform":"Copernicus journals","texts":217},{"platform":"Journal of Open Source Software","texts":209},{"platform":"Open Research Europe","texts":181},{"platform":"Peer Community In","texts":134},{"platform":"Nature Communications","texts":115},{"platform":"eLife","texts":82},{"platform":"Review Commons","texts":12},{"platform":"PubPeer","texts":10},{"platform":"F1000Research","texts":4},{"platform":"Qeios","texts":4},{"platform":"EMBO Press","texts":4}]},"agents":[{"name":"Methodology","role":"Audits study design, statistical power, and analytical choices against field-specific rigour standards (CONSORT, STROBE, PRISMA)."},{"name":"Formulas & Equations","role":"Verifies mathematical derivations, checks dimensional analysis, and flags algebraic errors."},{"name":"Originality","role":"Checks the manuscript's novelty claims against papers in our 7.9M-paper local library and flags possible overlap and self-plagiarism signals."},{"name":"Literature Coverage","role":"Evaluates citation completeness, missing seminal references and self-citation balance, with a live OpenAlex snapshot of the field (volume, top venues, peak year) for context."},{"name":"Reproducibility","role":"Inspects code availability, dataset accessibility, and sufficiency of methods detail for independent replication."},{"name":"Clarity & Language","role":"Assesses readability, structural flow, and adherence to scholarly writing norms."},{"name":"Figures & Tables","role":"Checks figure quality, caption completeness, and appropriateness of visual encodings."},{"name":"Research Gap","role":"Quotes the gap the manuscript says it fills, checks it against comparable published papers retrieved from our library (with their mined methods, data and results where we hold them), and scores whether that gap is real and addressed."},{"name":"Prior Publication","role":"A deterministic lookup, not a language model: fans out in parallel to 7 external sources — CrossRef, PubMed, arXiv, bioRxiv, medRxiv, Europe PMC and Unpaywall — and our local library to detect prior publication and duplicate submission."}],"fieldMix":[{"field":"machine learning and computer science","platforms":["OpenReview"],"texts":3033,"sharePct":54},{"field":"physics","platforms":["SciPost"],"texts":818,"sharePct":14},{"field":"life, health and earth sciences","platforms":["PeerJ","Copernicus journals","Peer Community In","eLife","Review Commons","F1000Research","EMBO Press"],"texts":755,"sharePct":13},{"field":"preprint reviews and comments across fields","platforms":["PREreview","PubPeer"],"texts":483,"sharePct":8},{"field":"multidisciplinary journals","platforms":["Open Research Europe","Nature Communications","Qeios"],"texts":300,"sharePct":5},{"field":"research software","platforms":["Journal of Open Source Software"],"texts":209,"sharePct":3}],"notes":"Aggregate counts only. counts.reviews counts distinct peer-review texts collected from the source platforms (referee reports, editors' decisions, public comments and authors' responses, each counted once however often it is stored); text extracted from papers in our library is not counted as reviews and is not part of this dataset. Per-review text is not served here, and the CC BY 4.0 licence covers only this summary: the texts remain under their source platforms' terms; contact editor@scienceaijournal.com for research collaboration access."}