{"id":"W7106831362","doi":"10.48448/azq0-f178","title":"MMLU-ProX: A Multilingual Benchmark for Advanced Large Language Model Evaluation","year":2025,"lang":"","type":"other","venue":"Open MIND","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Benchmark (surveying); Process (computing); Focus (optics); Quality (philosophy); Multilingualism","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":[0.01971897,0.003548026,0.001162062,0.004687625,0.001370919,0.005069442,0.004732781,0.002549776,0.0159351],"category_scores_gemma":[0.07534484,0.0008490058,0.00201767,0.003214217,0.001348093,0.007358257,0.00720245,0.002910733,0.009903771],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002387748,"about_ca_system_score_gemma":0.005039462,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00824308,"about_ca_topic_score_gemma":0.01088255,"domain_scores_codex":[0.9701887,0.01773009,0.003414149,0.002407283,0.005423648,0.0008361079],"domain_scores_gemma":[0.9634522,0.02024427,0.001352019,0.005058516,0.008589705,0.001303233],"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":[0.001837129,0.001942761,0.01350106,0.009237081,0.001033359,0.0006970167,0.003009496,0.05136092,0.01445296,0.03557462,0.419972,0.4473816],"study_design_scores_gemma":[0.001546093,0.002278042,0.01514691,0.002592023,0.0005275415,0.001235633,0.003149279,0.4357617,0.0510526,0.05924409,0.426777,0.0006890619],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.139542,0.01155135,0.4792667,0.004132707,0.002240388,0.004618502,0.08244415,0.1563572,0.1198471],"genre_scores_gemma":[0.2981818,0.001601216,0.4495607,0.001515187,0.0002247512,0.005469183,0.2190459,0.01436019,0.01004106],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01971897,"threshold_uncertainty_score":0.1042852,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05382144449479444,"score_gpt":0.4276749815402294,"score_spread":0.3738535370454349,"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."}}