{"id":"W4387582224","doi":"10.1016/j.imavis.2023.104838","title":"DiPS: Discriminative pseudo-label sampling with self-supervised transformers for weakly supervised object localization","year":2023,"lang":"en","type":"article","venue":"Image and Vision Computing","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":false,"ca_institutions":"Ericsson (Canada); École de Technologie Supérieure; Artificial Intelligence in Medicine (Canada)","funders":"Alliance de recherche numérique du Canada; Natural Sciences and Engineering Research Council of Canada; Mitacs","keywords":"Discriminative model; Artificial intelligence; Overfitting; Computer science; Pattern recognition (psychology); Transformer; Leverage (statistics); Classifier (UML); Novelty; Computer vision; Machine learning; Artificial neural network","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0002949167,0.0002245127,0.0002274879,0.000153079,0.0007412875,0.0002466109,0.0003237473,0.00005239459,0.000001468333],"category_scores_gemma":[0.00003294147,0.0001882462,0.00005043182,0.001101377,0.00006471302,0.0007820289,0.0001600478,0.0001304496,0.000009232601],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003366698,"about_ca_system_score_gemma":0.00003796444,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000006397428,"about_ca_topic_score_gemma":0.000004714378,"domain_scores_codex":[0.9983397,0.00005158566,0.0002996155,0.0006369702,0.0002376147,0.0004345074],"domain_scores_gemma":[0.998798,0.0005659483,0.00009200969,0.0002411936,0.0001819021,0.0001209193],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00006670246,0.0001619339,0.000287093,0.000240892,0.0000521361,0.00001093582,0.006605155,0.006789281,0.01875924,0.00513214,0.0006022439,0.9612923],"study_design_scores_gemma":[0.001043333,0.0002362196,0.0006118755,0.0001122001,0.00001904878,0.000009811632,0.000362194,0.9939997,0.00163409,0.001311712,0.0003909598,0.000268918],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04702498,0.00004273609,0.9504698,0.0008687385,0.00008232339,0.0007051694,0.000006604257,0.0006749101,0.0001247524],"genre_scores_gemma":[0.5828736,0.00006434276,0.4165581,0.0002776578,0.00008710216,0.00004429108,0.00004065861,0.00003226881,0.00002194586],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9872103,"threshold_uncertainty_score":0.7676455,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03188130234756244,"score_gpt":0.3336098988760882,"score_spread":0.3017285965285257,"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."}}