{"id":"W4417036181","doi":"10.48550/arxiv.2512.04072","title":"SkillFactory: Self-Distillation For Learning Cognitive Behaviors","year":2025,"lang":"","type":"preprint","venue":"arXiv (Cornell University)","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Alfred P. Sloan Foundation; Institute for Information and Communications Technology Promotion; Open Philanthropy Project; National Science Foundation","keywords":"Leverage (statistics); Reinforcement learning; Cognition; Initialization; Task (project management); Cognitive model; Priming (agriculture)","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","sts"],"consensus_categories":[],"category_scores_codex":[0.0006449759,0.0008226043,0.0007394335,0.0007078621,0.001411087,0.0003309925,0.002051545,0.0007394874,0.00009362825],"category_scores_gemma":[0.0004957231,0.001125003,0.0006284846,0.001540513,0.000226693,0.0005290541,0.00208693,0.001898269,0.000198812],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00076158,"about_ca_system_score_gemma":0.0008194153,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000908063,"about_ca_topic_score_gemma":0.00005085501,"domain_scores_codex":[0.9948598,0.0005844362,0.0005664204,0.002971868,0.0001949962,0.0008224534],"domain_scores_gemma":[0.9946091,0.001604763,0.001009735,0.001278282,0.001132064,0.0003660985],"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":[0.0001480321,0.0006412718,0.103816,0.0004010186,0.0004678559,0.0000333439,0.002816794,0.7526694,0.00004316565,0.1261993,0.00006141563,0.01270236],"study_design_scores_gemma":[0.001432162,0.0001859334,0.05230664,0.0003655449,0.0007977815,0.000002666012,0.000261718,0.9366078,0.00006959313,0.004900114,0.002057186,0.001012886],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3054175,0.00003215305,0.6884898,0.0001284135,0.0006058596,0.001841464,0.0001319364,0.0007428567,0.002610041],"genre_scores_gemma":[0.9855503,0.0001361405,0.007765801,0.00006786196,0.0001446589,0.00003177981,0.0002569718,0.00004683604,0.005999679],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.680724,"threshold_uncertainty_score":0.999889,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05909441059544795,"score_gpt":0.23556282801554,"score_spread":0.176468417420092,"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."}}