{"id":"W4413180594","doi":"10.1109/isscs66034.2025.11105702","title":"Learning Optimal Agent Behavior From a Synthetic Reasoning Action Dataset","year":2025,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Ontario Ministry of Research, Innovation and Science","keywords":"Computer science; Action (physics); Artificial intelligence; Machine learning","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002077023,0.001898866,0.0008082103,0.001185972,0.0006058779,0.001239564,0.00178103,0.001969317,0.00315745],"category_scores_gemma":[0.006801309,0.0005674079,0.001212762,0.0007333754,0.0008632019,0.001012589,0.0009052947,0.002249913,0.001035318],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002524208,"about_ca_system_score_gemma":0.001577633,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01732232,"about_ca_topic_score_gemma":0.02917262,"domain_scores_codex":[0.9989986,0.000382922,0.00004750768,0.0003520169,0.0001229428,0.00009605468],"domain_scores_gemma":[0.9958176,0.003039394,0.0002318199,0.000373672,0.0003295574,0.0002079337],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005266136,0.0005518876,0.008296259,0.0002361207,0.0001651125,0.0002214,0.0001100502,0.9226906,0.001520979,0.00263508,0.008255884,0.05479014],"study_design_scores_gemma":[0.0000411594,0.0000664254,0.0006326673,0.0000132695,0.00001269136,0.00001616314,0.00002451401,0.9954542,0.0006752809,0.002350019,0.0007055114,0.000008072352],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7910072,0.001917673,0.1747096,0.00230239,0.0003220224,0.000605995,0.01041414,0.006852428,0.0118686],"genre_scores_gemma":[0.9191374,0.0002222358,0.06383204,0.0003032877,0.00004561016,0.0002991115,0.01312194,0.0001607318,0.002877667],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01732232,"threshold_uncertainty_score":0.03444302,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02590422535685372,"score_gpt":0.3018162174335818,"score_spread":0.2759119920767281,"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."}}