{"id":"W2983647115","doi":"10.1109/ijcnn52387.2021.9533769","title":"Soft-Label Dataset Distillation and Text Dataset Distillation","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":31,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Distillation; Computer science; Code (set theory); Artificial intelligence; Sample (material); Pattern recognition (psychology); Image (mathematics); Task (project management); MNIST database; Machine learning; Data mining; Deep learning; Chromatography; Engineering","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":[0.002958115,0.002597383,0.001565923,0.002415835,0.001257567,0.002400216,0.003323961,0.002108629,0.01223208],"category_scores_gemma":[0.01561287,0.0009180292,0.002294029,0.002815782,0.001655221,0.006112138,0.004787696,0.004127816,0.005643093],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001683568,"about_ca_system_score_gemma":0.002066046,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003719966,"about_ca_topic_score_gemma":0.008734725,"domain_scores_codex":[0.9971535,0.0007392928,0.0002162244,0.001082087,0.0005722318,0.0002365825],"domain_scores_gemma":[0.9937232,0.002099063,0.0003331415,0.002932346,0.0006987242,0.0002136221],"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":[0.001337687,0.0006740743,0.005158173,0.001049313,0.0002959091,0.0003835837,0.0004761372,0.118075,0.04248253,0.02374332,0.07721219,0.7291121],"study_design_scores_gemma":[0.0002454269,0.0002524942,0.001461385,0.00007739318,0.0000640729,0.0002968439,0.0001679627,0.8839052,0.04314061,0.03585914,0.03443351,0.00009592337],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.05894656,0.001196767,0.8594255,0.001651362,0.0007873052,0.0008977334,0.01123749,0.06012781,0.005729442],"genre_scores_gemma":[0.1925547,0.0002971033,0.7620963,0.001203325,0.0002473219,0.001368077,0.03219669,0.003056865,0.006979503],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01223208,"threshold_uncertainty_score":0.04092032,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04822326090122638,"score_gpt":0.296521998139149,"score_spread":0.2482987372379227,"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."}}