{"id":"W4391767348","doi":"","title":"Do (colored) backgrounds matter? An experiment on artificially augmented ground truth for handwritten text recognition applied to historical manuscripts","year":2024,"lang":"en","type":"article","venue":"HAL (Le Centre pour la Communication Scientifique Directe)","topic":"Handwritten Text Recognition Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Colored; Ground truth; Computer science; Artificial intelligence; Natural language processing; Speech recognition; Pattern recognition (psychology); Sociology","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":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.003233152,0.0003690694,0.0003445398,0.0005156825,0.0004455402,0.001689734,0.00130554,0.0001872318,0.0002015596],"category_scores_gemma":[0.0002274516,0.0003856118,0.0001848173,0.000789434,0.00007700463,0.0006222929,0.0003358661,0.0002756626,0.000529828],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000758532,"about_ca_system_score_gemma":0.0001369314,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001780102,"about_ca_topic_score_gemma":0.0001368537,"domain_scores_codex":[0.9958003,0.001211421,0.0006361703,0.0012844,0.0005433671,0.0005243251],"domain_scores_gemma":[0.9960969,0.0009960664,0.0001628088,0.001494036,0.0008958191,0.0003543696],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001046231,0.001485309,0.00002203031,0.0001353067,0.00007939791,0.00001439947,0.008114287,0.000004419658,0.0462927,0.246822,0.02124415,0.6756814],"study_design_scores_gemma":[0.001760268,0.00005944528,0.001216361,0.002380096,0.00009369673,0.00005720769,0.0005118991,0.02331025,0.7375988,0.07536083,0.1556094,0.002041735],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04173371,0.0001960859,0.9212987,0.01036957,0.0006323973,0.001208531,0.00004304325,0.001218594,0.02329937],"genre_scores_gemma":[0.750468,0.00003740101,0.2425681,0.0008908761,0.00007950419,0.001209708,0.0003119383,0.00008141671,0.004353022],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7087343,"threshold_uncertainty_score":0.9998596,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03406623719093441,"score_gpt":0.2590868283488217,"score_spread":0.2250205911578873,"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."}}