{"id":"W6920575173","doi":"10.60692/r8a9d-3vs67","title":"Exploring the Limits of Simple Learners in Knowledge Distillation for Document Classification with DocBERT","year":2020,"lang":"en","type":"article","venue":"Greater South Information System","topic":"Handwritten Text Recognition Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; University of Waterloo","funders":"","keywords":"Distillation; Baseline (sea); Simple (philosophy); Document classification; Knowledge extraction; Knowledge acquisition; Natural language; Question answering","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001997009,0.001739583,0.0008313578,0.0006935582,0.0006055168,0.001642249,0.001827122,0.001608822,0.003486597],"category_scores_gemma":[0.006987485,0.0005196985,0.0003964486,0.0006445706,0.001012537,0.00633083,0.001996397,0.003434283,0.001700927],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009614162,"about_ca_system_score_gemma":0.00140999,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007632647,"about_ca_topic_score_gemma":0.01411384,"domain_scores_codex":[0.9994468,0.0001623175,0.00002958701,0.0001346981,0.0001262036,0.000100445],"domain_scores_gemma":[0.9977869,0.001403743,0.00008467274,0.0004114254,0.0002101108,0.0001031505],"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":[0.0008975235,0.0005289637,0.003317724,0.0004451841,0.0001881206,0.0002560952,0.0002941334,0.4149244,0.01728031,0.01434261,0.009809091,0.5377159],"study_design_scores_gemma":[0.00002776084,0.0001397243,0.000252955,0.00002856811,0.00002062367,0.0000462707,0.00005565668,0.9800863,0.009131209,0.008355075,0.001835896,0.00001991181],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4515185,0.006934475,0.4951689,0.00316663,0.000628449,0.0002250845,0.001096956,0.01903189,0.02222913],"genre_scores_gemma":[0.8807771,0.0007927453,0.1092334,0.0006607603,0.0001140029,0.0001218569,0.001143487,0.0004370634,0.006719578],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007632647,"threshold_uncertainty_score":0.01517642,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1506139091073968,"score_gpt":0.2681573197427652,"score_spread":0.1175434106353684,"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."}}