{"id":"W4390189903","doi":"10.1109/iccvw60793.2023.00465","title":"Confusing Large Models by Confusing Small Models","year":2023,"lang":"en","type":"article","venue":"","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Computer science; Robustness (evolution); Machine learning; Heuristics; Confusion; Artificial intelligence; Benchmark (surveying); Focus (optics); Upsampling; Image (mathematics)","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":[0.0007948943,0.0002355822,0.0002632031,0.0001919535,0.0004006459,0.0003451254,0.001065028,0.0001219116,0.00004773099],"category_scores_gemma":[0.00005482683,0.0002332141,0.00008604748,0.0008355689,0.00004403641,0.001310989,0.0009879901,0.0003364859,0.000150064],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000581207,"about_ca_system_score_gemma":0.00008987213,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001698348,"about_ca_topic_score_gemma":0.00001495312,"domain_scores_codex":[0.9977968,0.0001248548,0.000318138,0.0006345908,0.0003858714,0.0007396935],"domain_scores_gemma":[0.9988021,0.0001981884,0.0001107174,0.0006394697,0.0001039149,0.0001456095],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000002220984,0.0000189354,0.00002251596,0.000009578867,0.00001396113,0.00002256348,0.0008969952,0.6024911,0.000804441,0.3895919,0.002345084,0.003780724],"study_design_scores_gemma":[0.0004470221,0.00001372887,0.000006290999,0.0000205391,0.000005470035,0.000007055693,0.0001137751,0.9514135,0.0003939478,0.04566556,0.001624484,0.0002886345],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.008693817,0.00005050376,0.9606714,0.001084361,0.0004651547,0.0001624379,0.000004530655,0.001565949,0.02730181],"genre_scores_gemma":[0.8919317,0.00001939447,0.1028042,0.001233315,0.00009669615,0.00001037751,0.0000212205,0.0000434814,0.003839652],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8832379,"threshold_uncertainty_score":0.9510196,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04520528662869609,"score_gpt":0.2644070661826076,"score_spread":0.2192017795539115,"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."}}