{"id":"W3203851352","doi":"10.1016/j.neucom.2022.09.094","title":"On cropped versus uncropped training sets in tabular structure detection","year":2022,"lang":"en","type":"article","venue":"Neurocomputing","topic":"Handwritten Text Recognition Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":false,"ca_institutions":"Gnowit (Canada); University of Ottawa","funders":"","keywords":"Table (database); Intersection (aeronautics); Set (abstract data type); Computer science; Artificial intelligence; Precision and recall; Training set; Data mining; Machine learning; Geography; Cartography","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":[],"consensus_categories":[],"category_scores_codex":[0.0003053278,0.0001801656,0.0001836157,0.0003630023,0.0004284428,0.00012378,0.0006435133,0.00005041239,0.00004540367],"category_scores_gemma":[0.0001089089,0.0002091158,0.00006294975,0.0008715441,0.00001759,0.0001783159,0.000460149,0.000706836,0.000008163292],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001528472,"about_ca_system_score_gemma":0.00005195054,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002240892,"about_ca_topic_score_gemma":0.00002408796,"domain_scores_codex":[0.9979843,0.0003299775,0.0002923722,0.0005827592,0.0004168221,0.0003937461],"domain_scores_gemma":[0.9991477,0.000274751,0.0001378065,0.0003393741,0.00003588838,0.0000644793],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00008557156,0.00006718076,0.00008837117,0.00001353325,0.000009438985,0.0002073118,0.00171402,0.01056708,0.03344791,0.001504397,0.0001320777,0.9521631],"study_design_scores_gemma":[0.005797789,0.002372924,0.00432911,0.00009677696,0.00001453454,0.0002880063,0.0005669286,0.844107,0.1138404,0.02217386,0.005059661,0.001353055],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8571586,0.0000114314,0.1399496,0.0002476346,0.0009378407,0.0003185757,0.000004312609,0.0007463921,0.0006256712],"genre_scores_gemma":[0.9880496,7.324052e-7,0.01121372,0.0006188666,0.00005409522,0.00002719016,0.000004141387,0.00002248512,0.000009121763],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9508101,"threshold_uncertainty_score":0.8527493,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0270636309194266,"score_gpt":0.2630589935034269,"score_spread":0.2359953625840003,"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."}}