{"id":"W4308571509","doi":"10.5334/johd.97","title":"CREMMA Medii Aevi: Literary Manuscript Text Recognition in Latin","year":2023,"lang":"en","type":"article","venue":"Journal of Open Humanities Data","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Latin Americans; Computer science; Representation (politics); Natural language processing; Artificial intelligence; Segmentation; Medieval Latin; Information retrieval; Linguistics; History; Classics; Political science","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":["scholarly_communication","open_science"],"consensus_categories":[],"category_scores_codex":[0.001623098,0.0001256937,0.0002581006,0.0005204928,0.000107436,0.001511174,0.006673254,0.00005614461,0.0001283424],"category_scores_gemma":[0.0003150031,0.0001059043,0.00003254237,0.0005205374,0.00004553701,0.007549059,0.00351156,0.0004148528,0.00005919283],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006092148,"about_ca_system_score_gemma":0.000118873,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009399636,"about_ca_topic_score_gemma":0.0001052445,"domain_scores_codex":[0.9984253,0.0001318948,0.0005718817,0.000251868,0.0003971567,0.0002218648],"domain_scores_gemma":[0.998445,0.0001505288,0.0003954865,0.0007826054,0.0001774468,0.00004893091],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00006957789,0.0001396528,0.0004549309,0.0001556782,0.00004966218,0.002183281,0.007019273,0.000001268959,0.001032408,0.01124598,0.4570965,0.5205517],"study_design_scores_gemma":[0.002246122,0.0006883827,0.00476367,0.004090684,0.00005035517,0.0009240578,0.001847095,0.009224102,0.005362446,0.7734202,0.196238,0.00114495],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.4417606,0.09530307,0.2311466,0.07341314,0.02354151,0.007755398,0.002757518,0.005372938,0.1189492],"genre_scores_gemma":[0.2170743,0.001107055,0.7706659,0.004048653,0.001200692,0.00002446938,0.0006487126,0.00007700981,0.005153174],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7621742,"threshold_uncertainty_score":0.9995254,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.23807547872461,"score_gpt":0.3631402314952308,"score_spread":0.1250647527706208,"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."}}