{"id":"W4390872800","doi":"10.1109/iccv51070.2023.00190","title":"SAFARI: Versatile and Efficient Evaluations for Robustness of Interpretability","year":2023,"lang":"en","type":"article","venue":"","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":23,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"University of Warwick; Transport Canada","keywords":"Interpretability; Robustness (evolution); Computer science; Machine learning; Artificial intelligence; Probabilistic logic; Bespoke; Data mining; Trustworthiness","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001118907,0.00006657286,0.0001124964,0.0001062439,0.00009461317,0.0000457982,0.0003363146,0.00003037074,0.00003406465],"category_scores_gemma":[0.0005353099,0.00005999504,0.0000428991,0.0005415515,0.00007637348,0.0001557847,0.0002155827,0.00003357232,0.00002250589],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002745699,"about_ca_system_score_gemma":0.00006098887,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004496841,"about_ca_topic_score_gemma":0.00004139931,"domain_scores_codex":[0.9990959,0.00005562424,0.000222867,0.0002682684,0.0001719639,0.0001853851],"domain_scores_gemma":[0.9986559,0.0006596175,0.00004741108,0.0003815974,0.0002088112,0.00004673156],"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":[0.00005485387,0.0002562476,0.001133004,0.0001840962,0.00004095895,0.000001325976,0.009057344,0.4778358,0.006641695,0.3812651,0.002673676,0.1208559],"study_design_scores_gemma":[0.00006358372,0.00006347204,0.001228538,0.000008702758,0.000004485884,5.353396e-7,0.0004433203,0.972576,0.02150485,0.003861928,0.0001782307,0.00006641018],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2298944,0.0000126935,0.7680708,0.0006497017,0.0002255761,0.0003173623,0.000003636638,0.000116327,0.0007094423],"genre_scores_gemma":[0.9801984,0.00000211118,0.01950766,0.0000269887,0.00001236665,0.00005348692,0.000001991356,0.000004015785,0.0001930262],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7503039,"threshold_uncertainty_score":0.2446527,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04740619807341593,"score_gpt":0.3414541511880035,"score_spread":0.2940479531145875,"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."}}