{"id":"W4389518309","doi":"10.18653/v1/2023.arabicnlp-1.6","title":"TARJAMAT: Evaluation of Bard and ChatGPT on Machine Translation of Ten Arabic Varieties","year":2023,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":24,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","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":"Arabic; Constraint (computer-aided design); Computer science; Linguistics; Modern Standard Arabic; Natural language processing; Machine translation; Artificial intelligence; Engineering","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0008122068,0.00005863862,0.0000988437,0.0001611121,0.000023481,0.00001798323,0.0001826949,0.00004075672,0.00001178287],"category_scores_gemma":[0.00005048806,0.0000444285,0.00001933961,0.0003551964,0.00002941286,0.000212302,0.00004050127,0.00004921668,0.000001856469],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000008693132,"about_ca_system_score_gemma":0.00002547316,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004803496,"about_ca_topic_score_gemma":0.000008997758,"domain_scores_codex":[0.9991972,0.00005432496,0.0001478828,0.0001402013,0.0003918639,0.00006854747],"domain_scores_gemma":[0.9995313,0.00006935958,0.00006959733,0.0001795482,0.0001356171,0.0000146028],"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":[0.00002089226,0.00003917907,0.0003834626,0.000147163,0.00001770791,9.379673e-7,0.003689093,0.0002113844,0.02905001,0.1010098,0.0001985992,0.8652318],"study_design_scores_gemma":[0.0003983615,0.0002101854,0.00569942,0.00007634193,0.00002232099,0.000002612473,0.00003754267,0.5343422,0.1967605,0.2622837,0.00003294063,0.0001338512],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2190568,0.00422351,0.7672214,0.003145953,0.0001614316,0.000653533,0.000008706656,0.001077982,0.004450635],"genre_scores_gemma":[0.9132192,0.00001850568,0.0866567,0.00003185944,0.000006016119,0.00000657414,0.000004565704,0.000003313189,0.00005334229],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8650979,"threshold_uncertainty_score":0.1811742,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03261906517732031,"score_gpt":0.3079726270094847,"score_spread":0.2753535618321644,"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."}}