{"id":"W3127215335","doi":"10.1609/aaai.v35i9.16969","title":"Show, Attend and Distill: Knowledge Distillation via Attention-based Feature Matching","year":2021,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":145,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Distillation; Computer science; Feature (linguistics); Matching (statistics); Artificial intelligence; Machine learning; Selection (genetic algorithm); Control (management); Knowledge transfer; Mathematics; Knowledge management; Statistics","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":[0.0003505324,0.0001950953,0.0002085321,0.0000904593,0.0003139204,0.000421603,0.0006349404,0.00008585895,0.00004561145],"category_scores_gemma":[0.0002987928,0.0001588456,0.00009637892,0.0005927121,0.0001733022,0.0003452325,0.0002472832,0.0003024831,0.00004134967],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003843174,"about_ca_system_score_gemma":0.0001024866,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001029467,"about_ca_topic_score_gemma":0.00002091658,"domain_scores_codex":[0.9985214,0.0000343214,0.0003405586,0.0005014382,0.0003531342,0.0002491488],"domain_scores_gemma":[0.998612,0.0001463839,0.0002727915,0.0002248201,0.0006457041,0.00009833711],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002662993,0.0001160318,0.001132203,0.00008131869,0.00001453707,0.000001512701,0.00147632,0.0001585341,0.06416903,0.7478313,0.00007824117,0.1849144],"study_design_scores_gemma":[0.00009744433,0.0001367238,0.009997787,0.0005331121,0.00002958656,0.00001964035,0.001205457,0.7499199,0.1046961,0.1321152,0.0007616007,0.0004874478],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1840732,0.0001255224,0.7916111,0.009221938,0.0007236637,0.0003489364,0.000006572832,0.0001745888,0.01371452],"genre_scores_gemma":[0.9905658,0.00001487938,0.00847979,0.0001695915,0.00005318475,0.00001022166,0.000002510855,0.0000116513,0.0006923125],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8064926,"threshold_uncertainty_score":0.6477535,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04860094291411718,"score_gpt":0.2885622224527176,"score_spread":0.2399612795386004,"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."}}