{"id":"W4414769236","doi":"10.31274/jlsc.19779","title":"Evaluating Multilingual Metadata Quality in Crossref","year":2025,"lang":"en","type":"article","venue":"Journal of Librarianship and Scholarly Communication","topic":"Intellectual Property Law","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University; Trinity College","funders":"","keywords":"Metadata; Multilingualism; Categorization; Publishing; Sample (material); Quality (philosophy)","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.01194403,0.00007327581,0.0002116348,0.000219867,0.0005600664,0.001901494,0.000890421,0.0001280306,0.00004154198],"category_scores_gemma":[0.01060605,0.00006360748,0.00005407779,0.0004545759,0.0002215515,0.008162004,0.0001702513,0.0008648031,0.00000191839],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000951371,"about_ca_system_score_gemma":0.0007087184,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001155578,"about_ca_topic_score_gemma":0.001114192,"domain_scores_codex":[0.9957504,0.002955878,0.0006733579,0.000101729,0.0003640897,0.0001545996],"domain_scores_gemma":[0.997548,0.001379921,0.0003137282,0.0003559271,0.0003366655,0.00006575122],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"observational","study_design_scores_codex":[0.0006629268,0.0004033546,0.1637668,0.00006074949,0.0001515924,0.000006882626,0.1374979,0.0000512866,0.006058857,0.4856084,0.0009027087,0.2048285],"study_design_scores_gemma":[0.005965448,0.0004195012,0.6090185,0.001465346,0.0001735261,0.00001359792,0.06474032,0.0009641008,0.003894662,0.2057448,0.1068892,0.0007110268],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9581934,0.004396765,0.0004450639,0.00928561,0.0001510975,0.0001485562,0.000001583004,0.00001385818,0.02736408],"genre_scores_gemma":[0.9897357,0.0005385016,0.007874686,0.0004878543,0.00004945411,0.000001852884,0.000001902915,0.000004922499,0.001305073],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4452517,"threshold_uncertainty_score":0.9991346,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2579584788334103,"score_gpt":0.4846089632726107,"score_spread":0.2266504844392003,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}