{"id":"W6891769059","doi":"10.48448/x2c7-r841","title":"An Empirical Study of Multilingual Reasoning Distillation for Question Answering","year":2024,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Correctness; Distillation; Variety (cybernetics); Empirical research; Question answering; Case-based reasoning; Empirical evidence","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.00198557,0.0003358301,0.0004082151,0.001275011,0.0001414336,0.0001672978,0.0005548986,0.0001825426,0.00007586109],"category_scores_gemma":[0.0007075749,0.0003086982,0.0000554407,0.001292586,0.0005230842,0.0002610522,0.0001209126,0.0002501797,0.0001901889],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002979056,"about_ca_system_score_gemma":0.0003996955,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006877513,"about_ca_topic_score_gemma":0.00272263,"domain_scores_codex":[0.9969073,0.00009136787,0.0004901455,0.001096393,0.0009830708,0.0004317547],"domain_scores_gemma":[0.9984561,0.00007032483,0.0003970503,0.0006359734,0.0002779486,0.0001626146],"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":[0.001666992,0.02480213,0.1376249,0.005530699,0.001398673,0.0002413768,0.1021982,0.0491337,0.1925131,0.0116898,0.1959277,0.2772728],"study_design_scores_gemma":[0.002152117,0.003354211,0.004477008,0.00197141,0.0005106211,0.00002078377,0.008520492,0.9479395,0.001180012,0.0009632545,0.02737498,0.001535667],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8076129,0.001329043,0.05687433,0.000098867,0.006504257,0.01163616,0.001521298,0.00598509,0.1084381],"genre_scores_gemma":[0.9613578,0.000001947487,0.02166101,0.00000823991,0.0005918693,0.00004413326,0.0001065243,0.0007500311,0.01547842],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8988057,"threshold_uncertainty_score":0.9999365,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0436910017129451,"score_gpt":0.4279754972270292,"score_spread":0.3842844955140841,"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."}}