{"id":"W3176169354","doi":"10.18653/v1/2021.acl-long.551","title":"ARBERT & MARBERT: Deep Bidirectional Transformers for Arabic","year":2021,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":352,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Social Sciences and Humanities Research Council of Canada; Natural Sciences and Engineering Research Council of Canada; Compute Canada","keywords":"Arabic; Transformer; Computer science; Natural language processing; Linguistics; Artificial intelligence; Speech recognition; Electrical engineering; Engineering; Philosophy; Voltage","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":[0.000126079,0.00008159246,0.00009660102,0.00004338017,0.00008761097,0.00007224285,0.0002029347,0.00004601233,0.0001809896],"category_scores_gemma":[0.00003142933,0.00007574941,0.00009876957,0.0001853638,0.00001086594,0.0002296935,0.00003546252,0.00006172847,0.00004874122],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003276566,"about_ca_system_score_gemma":0.00009967976,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002375183,"about_ca_topic_score_gemma":0.0003118924,"domain_scores_codex":[0.9991296,0.00001642621,0.0001520258,0.0003280997,0.0001583641,0.0002154632],"domain_scores_gemma":[0.9995211,0.0000883754,0.0000168786,0.0002128907,0.00009037137,0.00007039667],"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":[0.00001526919,0.0001708915,0.0007777142,0.00007093592,0.0001035584,0.00001212666,0.0009052477,0.001852788,0.00761759,0.2705685,0.005342091,0.7125633],"study_design_scores_gemma":[0.0008176357,0.00003162233,0.0007120128,0.00001632836,0.00001420018,0.000111935,0.00004404889,0.5923514,0.01304568,0.0262104,0.3662167,0.0004280714],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.003081512,0.0001191326,0.9710876,0.003222484,0.0005850106,0.00009310334,9.496487e-7,0.0001306743,0.02167959],"genre_scores_gemma":[0.1499771,0.00002857494,0.8246242,0.00202831,0.0002046876,0.00006544343,0.00001126004,0.00001389396,0.02304654],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.7121352,"threshold_uncertainty_score":0.3088971,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02856709349934125,"score_gpt":0.2625137650090602,"score_spread":0.233946671509719,"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."}}