{"id":"W4308327073","doi":"10.1007/s10994-022-06263-z","title":"Adversarial examples for extreme multilabel text classification","year":2022,"lang":"fi","type":"article","venue":"Aaltodoc (Aalto University)","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":false,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"Aalto-Yliopisto; Academy of Finland","keywords":"Adversarial system; Robustness (evolution); Computer science; Artificial intelligence; Similarity (geometry); Categorization; Text categorization; Machine learning; Pattern recognition (psychology); Data mining; 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":["metaepi_narrow","sts","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001061543,0.0005545603,0.0005795044,0.0009547157,0.003115285,0.0002233639,0.00329764,0.0002327635,0.0009764695],"category_scores_gemma":[0.0003772828,0.0007612739,0.0004160052,0.002122804,0.0002871489,0.001337084,0.002612241,0.001076416,0.0001381395],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001431291,"about_ca_system_score_gemma":0.0008141759,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006185675,"about_ca_topic_score_gemma":0.00008569822,"domain_scores_codex":[0.9948354,0.0009676291,0.0005447111,0.001647887,0.0009131492,0.001091179],"domain_scores_gemma":[0.9962503,0.001022045,0.0007549674,0.00130555,0.0003238993,0.0003431871],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001493621,0.000924962,0.002055378,0.0002339096,0.0003995466,0.0003388343,0.006873778,0.07635129,0.002851511,0.7753077,0.01306944,0.1201],"study_design_scores_gemma":[0.003575361,0.0004349986,0.001270582,0.00003144604,0.0001713069,0.00001980329,0.003059141,0.3145114,0.00005061308,0.0007050519,0.6754225,0.0007477617],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00973702,0.0003567575,0.9541851,0.004643173,0.006499959,0.002042472,0.0003012761,0.0006496853,0.02158459],"genre_scores_gemma":[0.8851511,0.000114782,0.05893964,0.0005302807,0.001021838,0.00002901051,0.0002211,0.0001297312,0.0538625],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8952454,"threshold_uncertainty_score":0.9999368,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07888814776326718,"score_gpt":0.2577371208845423,"score_spread":0.1788489731212751,"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."}}