{"id":"W7131097258","doi":"10.1109/iccvw69036.2025.00708","title":"Task-Specific Generative Dataset Distillation with Difficulty-Guided Sampling","year":2025,"lang":"","type":"article","venue":"","topic":"Machine Learning and Data Classification","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Preprocessor; Task (project management); Distillation; Matching (statistics); Sampling (signal processing); Transformation (genetics); Downstream (manufacturing); Focus (optics)","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":[0.001890273,0.001433882,0.001367512,0.0009282115,0.0005798606,0.001466926,0.002675995,0.001333877,0.006091987],"category_scores_gemma":[0.008658418,0.0006141233,0.001392656,0.001270249,0.001041912,0.002634123,0.003890166,0.003132302,0.001916268],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00094624,"about_ca_system_score_gemma":0.001281904,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002060374,"about_ca_topic_score_gemma":0.005177111,"domain_scores_codex":[0.9990269,0.0002792812,0.00005577974,0.000324892,0.0001947207,0.00011848],"domain_scores_gemma":[0.9973903,0.001163751,0.0001550317,0.0009138834,0.0002271588,0.0001497911],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0008877045,0.0005727111,0.004522192,0.000525202,0.0002357528,0.0003890246,0.0004671339,0.3880462,0.03353269,0.04880358,0.02926687,0.4927509],"study_design_scores_gemma":[0.000078373,0.0000807889,0.0003434283,0.00001673615,0.00001824676,0.00009852766,0.00003232864,0.9626788,0.008770063,0.02461065,0.003245476,0.00002665096],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03202894,0.0002981319,0.9568373,0.0003804096,0.000137742,0.0003082428,0.000826336,0.006896133,0.002286638],"genre_scores_gemma":[0.4194424,0.0002051905,0.5656182,0.0008170008,0.000165758,0.001054932,0.006148615,0.001717102,0.004830865],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006091987,"threshold_uncertainty_score":0.02037972,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04802980133914164,"score_gpt":0.319385498422392,"score_spread":0.2713556970832504,"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."}}