{"id":"W4401042975","doi":"10.18653/v1/2024.naacl-long.420","title":"Interplay of Machine Translation, Diacritics, and Diacritization","year":2024,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Alliance de recherche numérique du Canada; Social Sciences and Humanities Research Council of Canada; Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs","keywords":"Machine translation; Computer science; Translation (biology); Artificial intelligence; Natural language processing; Biology; Messenger RNA","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002488988,0.0006702208,0.0007717972,0.001472952,0.001831658,0.007224312,0.0007932928,0.001183161,0.01153114],"category_scores_gemma":[0.01623089,0.0007491912,0.0004746135,0.001919877,0.002261838,0.01002614,0.003100081,0.002060812,0.005550535],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000912429,"about_ca_system_score_gemma":0.001177457,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001209998,"about_ca_topic_score_gemma":0.002020941,"domain_scores_codex":[0.9982978,0.000894785,0.0001037303,0.0003081496,0.0002627236,0.0001328208],"domain_scores_gemma":[0.9907826,0.005425137,0.0005786934,0.001639849,0.001324925,0.0002487171],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0008109795,0.0003043451,0.004966255,0.001397281,0.0001568516,0.0007727141,0.004127576,0.007588468,0.03213238,0.5475087,0.03306025,0.3671743],"study_design_scores_gemma":[0.00008788173,0.0001487477,0.004374822,0.0001888713,0.0001655921,0.0007289255,0.001556474,0.06957316,0.02688387,0.7969898,0.09918255,0.0001192573],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.2884431,0.02799273,0.3446743,0.02631121,0.004706124,0.0002373813,0.001835559,0.006509746,0.29929],"genre_scores_gemma":[0.9231995,0.003396295,0.0546081,0.0006913461,0.0008034929,0.00006603486,0.000711954,0.001460343,0.01506282],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01153114,"threshold_uncertainty_score":0.03857553,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006933722539970036,"score_gpt":0.2855556771529202,"score_spread":0.2786219546129502,"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."}}