{"id":"W6910736779","doi":"10.48550/arxiv.2304.14044","title":"Large Scale Genealogical Information Extraction From Handwritten Quebec Parish Records","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Authorship Attribution and Profiling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Workflow; Consistency (knowledge bases); Information extraction; Sample (material); Scale (ratio); Process (computing)","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0005522112,0.0002579924,0.0002850684,0.0002441732,0.0002902081,0.000336765,0.0011001,0.0005794946,0.00009569084],"category_scores_gemma":[0.00007301982,0.0002967361,0.0001996611,0.0005825062,0.0000447322,0.001372605,0.001424768,0.0008268404,0.001029618],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003145791,"about_ca_system_score_gemma":0.0002101791,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00351938,"about_ca_topic_score_gemma":0.002653502,"domain_scores_codex":[0.9981968,0.0002307687,0.000317991,0.0007227666,0.0001467655,0.0003848959],"domain_scores_gemma":[0.9983157,0.000145029,0.0003470841,0.0008157648,0.0001875118,0.0001889208],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005479413,0.0008343904,0.1362909,0.0004959941,0.0007891128,0.001024259,0.01551232,0.2250242,0.0002737484,0.5500284,0.03573816,0.03344063],"study_design_scores_gemma":[0.000650436,0.00003609796,0.04194003,0.00009944123,0.00006499706,0.00000269977,0.0005086713,0.8562962,0.0002928392,0.08675326,0.01267738,0.0006779023],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1734466,0.00001833629,0.8229377,0.0002749681,0.001426803,0.0002099753,0.0001168289,0.0006723039,0.0008965163],"genre_scores_gemma":[0.9930508,0.0001457147,0.002650079,0.0001865484,0.0001759934,0.000002417886,0.000606987,0.00001160559,0.003169889],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8202876,"threshold_uncertainty_score":0.9999485,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1030290525406084,"score_gpt":0.2252851522090067,"score_spread":0.1222560996683982,"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."}}