{"id":"W4231122779","doi":"10.18653/v1/2021.emnlp-main.288","title":"Automated Generation of Accurate & Fluent Medical X-ray Reports","year":2021,"lang":"en","type":"article","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","topic":"Topic Modeling","field":"Computer Science","cited_by":38,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Fluency; Embedding; Generator (circuit theory); Transformer; Natural language processing; Artificial intelligence; Information retrieval; Linguistics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002156911,0.001502583,0.0007380606,0.001488018,0.0001871821,0.001144839,0.001704318,0.001013317,0.005209121],"category_scores_gemma":[0.01298163,0.0005293462,0.001158099,0.0006225577,0.0004265469,0.001269204,0.00169177,0.0008990154,0.003587427],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004288535,"about_ca_system_score_gemma":0.0008585897,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006545288,"about_ca_topic_score_gemma":0.0008831666,"domain_scores_codex":[0.9983838,0.0005927716,0.0001419191,0.0003954493,0.0004050018,0.00008110123],"domain_scores_gemma":[0.9920467,0.005440183,0.0005026309,0.0009599765,0.0008777683,0.0001728486],"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":[0.00121249,0.000375575,0.006043965,0.001572951,0.0002024853,0.00293525,0.000895316,0.09937468,0.06396499,0.009044558,0.04680929,0.7675684],"study_design_scores_gemma":[0.0002356734,0.000318624,0.002150985,0.0001321654,0.0001442515,0.001921802,0.0001877435,0.8355345,0.1178633,0.01776593,0.0236655,0.00007956575],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0452156,0.0006484048,0.9102278,0.0006249353,0.0002325203,0.0004003783,0.003748437,0.03680521,0.002096687],"genre_scores_gemma":[0.2848516,0.0005410219,0.6947877,0.0003286911,0.0002060666,0.0004142154,0.01250799,0.00266147,0.003701208],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005209121,"threshold_uncertainty_score":0.01742625,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09840819144765348,"score_gpt":0.4311545131288209,"score_spread":0.3327463216811675,"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."}}