{"id":"W4389519573","doi":"10.18653/v1/2023.findings-emnlp.98","title":"Dolphin: A Challenging and Diverse Benchmark for Arabic NLG","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":7,"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":"Computer science; Benchmark (surveying); Automatic summarization; Natural language processing; Modular design; Artificial intelligence; Set (abstract data type); Modern Standard Arabic; Generalization; Arabic; Linguistics; Programming language","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.008658279,0.002983898,0.001277803,0.005598159,0.002852117,0.00379057,0.003626246,0.003100671,0.009809377],"category_scores_gemma":[0.03006265,0.0005124203,0.00147029,0.005429951,0.001720277,0.005207325,0.005031507,0.002600463,0.007268904],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002942663,"about_ca_system_score_gemma":0.002974224,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01780516,"about_ca_topic_score_gemma":0.020123,"domain_scores_codex":[0.9923463,0.003675358,0.0007330459,0.001392191,0.001446937,0.0004062096],"domain_scores_gemma":[0.9871157,0.006424061,0.0004000946,0.002826584,0.002573991,0.0006596261],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001548311,0.001232325,0.01069463,0.003814702,0.0005234099,0.0009613642,0.002215955,0.1106084,0.008215095,0.02067578,0.4732822,0.3662279],"study_design_scores_gemma":[0.0008992854,0.000766024,0.01181408,0.0008174597,0.0001894475,0.001514195,0.003175054,0.502055,0.02705516,0.04599812,0.4054605,0.0002556455],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.33376,0.01852143,0.2077373,0.01059066,0.004414169,0.004341976,0.201858,0.08501227,0.1337642],"genre_scores_gemma":[0.3250145,0.002472298,0.2121308,0.00202547,0.0005265559,0.002883355,0.4318538,0.006253694,0.01683944],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01780516,"threshold_uncertainty_score":0.0457899,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0487968856808134,"score_gpt":0.2677313356933869,"score_spread":0.2189344500125735,"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."}}