{"id":"W4404672046","doi":"10.1007/978-3-031-73024-5_1","title":"How to Train the Teacher Model for Effective Knowledge Distillation","year":2024,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Technology-Enhanced Education Studies","field":"Social Sciences","cited_by":6,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Distillation; Artificial intelligence","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":[],"consensus_categories":[],"category_scores_codex":[0.001051498,0.0002524602,0.0002620099,0.000369055,0.0007021436,0.0003288745,0.000892006,0.0002683649,0.000004132387],"category_scores_gemma":[0.0005630674,0.0001842783,0.00009725573,0.0005306408,0.001385764,0.000135128,0.0002750188,0.0004622274,0.00002456679],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007952171,"about_ca_system_score_gemma":0.0005147966,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002495025,"about_ca_topic_score_gemma":0.002243606,"domain_scores_codex":[0.9982451,0.00002552426,0.0001689411,0.0007754869,0.0003658664,0.0004190801],"domain_scores_gemma":[0.9983788,0.0008410315,0.00008824063,0.0003713503,0.0002508087,0.00006982073],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00000548095,0.000009760614,0.00000551535,0.00002041859,0.00001566643,4.617189e-7,0.05172503,0.006756304,0.00003470002,0.1141661,0.0007029197,0.8265576],"study_design_scores_gemma":[0.00007845514,0.00006848336,0.00009297773,0.0001900329,0.0000260286,9.764592e-7,0.00004188732,0.1928964,0.0001106334,0.7589894,0.04711842,0.0003863179],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0001636285,0.0007475704,0.9670234,0.01949762,0.001727044,0.001537765,0.00001287481,0.0003086772,0.008981396],"genre_scores_gemma":[0.9157532,0.00002304008,0.04163103,0.0004153442,0.001422276,0.0003231032,0.000002230653,0.00003927004,0.04039051],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9253924,"threshold_uncertainty_score":0.7514649,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03277099080217485,"score_gpt":0.338730442324303,"score_spread":0.3059594515221282,"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."}}