{"id":"W4385573558","doi":"10.18653/v1/2022.wanlp-1.7","title":"A Benchmark Study of Contrastive Learning for Arabic Social Meaning","year":2022,"lang":"en","type":"article","venue":"","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Meaning (existential); Benchmark (surveying); Arabic; Computer science; Natural language processing; Linguistics; Artificial intelligence; Contrastive analysis; Psychology; Philosophy; Geology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002926202,0.00064703,0.0003466101,0.001351354,0.0009354518,0.001405342,0.001051929,0.0006568009,0.005637991],"category_scores_gemma":[0.01757451,0.000138965,0.0003338795,0.00111325,0.0006487516,0.003079958,0.001784803,0.0009627435,0.001259927],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001089798,"about_ca_system_score_gemma":0.0004696961,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004305091,"about_ca_topic_score_gemma":0.004315692,"domain_scores_codex":[0.9983828,0.0009678073,0.0001015041,0.0002035578,0.0002851112,0.00005916451],"domain_scores_gemma":[0.9853107,0.0114558,0.0002512146,0.001206712,0.001423637,0.000351999],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00260998,0.003416914,0.01493009,0.001252034,0.0002017916,0.0004197766,0.003989266,0.02760782,0.01222557,0.02799678,0.02038602,0.8849639],"study_design_scores_gemma":[0.000729588,0.005167107,0.04297113,0.000403803,0.0002369891,0.0008576789,0.009468509,0.7040729,0.05206754,0.0839458,0.09991868,0.0001603033],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8704358,0.004488595,0.06175617,0.001346949,0.0002917782,0.0007201229,0.002331957,0.001141501,0.05748713],"genre_scores_gemma":[0.9363419,0.0005469869,0.05536015,0.0001332942,0.00008056896,0.0002434674,0.002450862,0.0001111775,0.004731437],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.005637991,"threshold_uncertainty_score":0.018861,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02750089901426163,"score_gpt":0.2749420696812646,"score_spread":0.247441170667003,"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."}}