{"id":"W3091508658","doi":"10.1109/ijcnn48605.2020.9207309","title":"Cross-representation transferability of adversarial attacks: From spectrograms to audio waveforms","year":2020,"lang":"en","type":"article","venue":"Espace ÉTS (ETS)","topic":"Digital Media Forensic Detection","field":"Computer Science","cited_by":17,"is_retracted":false,"has_abstract":true,"ca_institutions":"École de Technologie Supérieure; Université du Québec à Montréal; McGill University","funders":"","keywords":"Spectrogram; Computer science; Transferability; Speech recognition; Convolutional neural network; Waveform; Artificial intelligence; Adversarial system; Fourier transform; Pattern recognition (psychology); Mathematics; Machine learning; Telecommunications","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":[0.000156601,0.0001821979,0.0002935247,0.00006318188,0.00005317239,0.0002223785,0.000622267,0.00008279066,0.00005894961],"category_scores_gemma":[0.0003460734,0.0001680333,0.0001547323,0.0007748331,0.000100403,0.001059636,0.00017387,0.0001541255,0.0002152513],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008720023,"about_ca_system_score_gemma":0.0000644798,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002982701,"about_ca_topic_score_gemma":0.0001642993,"domain_scores_codex":[0.9981197,0.00004917327,0.0003373396,0.0006314439,0.0005575341,0.0003048475],"domain_scores_gemma":[0.9988112,0.0001164552,0.00008879176,0.0005733635,0.0001188753,0.0002913578],"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":[0.001903367,0.0006203762,0.03386505,0.0001322081,0.0003109112,0.00007899303,0.05348637,0.006857268,0.07744204,0.009719482,0.003079598,0.8125044],"study_design_scores_gemma":[0.005046999,0.002812013,0.21324,0.00009166032,0.00006480147,0.00001220553,0.0007715374,0.04122338,0.707615,0.01042361,0.01744599,0.001252755],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7431614,0.000009345932,0.2492177,0.004263043,0.001109061,0.0004072865,0.00002428865,0.0002239229,0.001583964],"genre_scores_gemma":[0.9891341,0.000001789115,0.01010848,0.0003011545,0.0003123124,0.00001731229,0.0000139925,0.00001421085,0.0000966778],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8112516,"threshold_uncertainty_score":0.6852199,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01899236776660037,"score_gpt":0.2721492946560581,"score_spread":0.2531569268894577,"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."}}