{"id":"W4389518321","doi":"10.18653/v1/2023.arabicnlp-1.25","title":"Arabic Fine-Grained Entity Recognition","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia Hospital","funders":"","keywords":"Computer science; Natural language processing; Named-entity recognition; Arabic; Artificial intelligence; Entity linking; Modern Standard Arabic; Information retrieval; Linguistics; Knowledge base","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0001833084,0.00004282967,0.0000471504,0.00007422093,0.00004627404,0.00006045376,0.0002590769,0.00002274908,0.00008045443],"category_scores_gemma":[0.00004568119,0.00003877867,0.00002831741,0.000359802,0.000005655323,0.0002353401,0.0001220221,0.00004292123,0.001583389],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001040244,"about_ca_system_score_gemma":0.00001534301,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003492648,"about_ca_topic_score_gemma":0.00003074676,"domain_scores_codex":[0.9994479,0.00001744746,0.00009034255,0.0001906882,0.0001160044,0.0001376621],"domain_scores_gemma":[0.9996272,0.00003201065,0.00001607658,0.0002597433,0.00003142696,0.00003352545],"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":[0.000001709778,0.00003323301,0.001432622,0.00002122319,0.0000116427,0.0000289787,0.0004856197,0.0002691684,0.002181462,0.06697605,0.01683143,0.9117269],"study_design_scores_gemma":[0.0002448141,0.00002217854,0.00408595,0.00001132838,0.000002172437,0.000005124721,0.00001566614,0.9117165,0.002598993,0.07601031,0.005115332,0.0001716646],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1097526,0.000004175906,0.8792073,0.001873128,0.0003457338,0.00005494089,5.518801e-7,0.0007260214,0.008035517],"genre_scores_gemma":[0.9304193,0.000005873326,0.06146262,0.0003703164,0.0001017632,0.00001274405,0.00000847094,0.000004533088,0.007614355],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9115552,"threshold_uncertainty_score":0.999194,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0572780891113574,"score_gpt":0.2594864063940251,"score_spread":0.2022083172826677,"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."}}