{"id":"W6966508981","doi":"10.48448/3vkx-ch90","title":"PromptMix: A Class Boundary Augmentation Method for Large Language Model Distillation | VIDEO","year":2023,"lang":"en","type":"other","venue":"Open MIND","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo; McGill University; Minnow Environmental (Canada)","funders":"","keywords":"Correctness; Classifier (UML); Class (philosophy); Language model; False positive paradox; Labeled data; Training set; Transfer of learning; Code (set theory)","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":[0.001120227,0.00187778,0.0009118193,0.0009963713,0.0007432749,0.001209933,0.002299966,0.001513224,0.00943653],"category_scores_gemma":[0.004893683,0.0005782053,0.001190638,0.0007768891,0.0008753901,0.002798664,0.003051474,0.003460091,0.005617892],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007996392,"about_ca_system_score_gemma":0.001053616,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004023535,"about_ca_topic_score_gemma":0.007675206,"domain_scores_codex":[0.9990709,0.0002164124,0.00003961881,0.0003622234,0.0002156314,0.00009504839],"domain_scores_gemma":[0.9988993,0.0004870672,0.0000723889,0.0003054093,0.000167833,0.00006797014],"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":[0.0005889898,0.0002690032,0.0009352738,0.0003132006,0.00006632059,0.0002492427,0.00029372,0.03790338,0.03365283,0.008450727,0.05608813,0.8611892],"study_design_scores_gemma":[0.00007105015,0.0001301685,0.0004092914,0.00003860502,0.00001457412,0.0001182704,0.00009725279,0.9389862,0.02574381,0.01436995,0.01997615,0.00004462534],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01757754,0.0006323086,0.929747,0.0004553288,0.0004078436,0.0002352115,0.002599742,0.04511452,0.003230545],"genre_scores_gemma":[0.1782147,0.0003219357,0.7940975,0.000679038,0.000286931,0.0007661734,0.01258654,0.003260082,0.009787179],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.00943653,"threshold_uncertainty_score":0.03156829,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05326630746925805,"score_gpt":0.4116189090974773,"score_spread":0.3583526016282193,"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."}}