{"id":"W4407830460","doi":"10.1109/jbhi.2025.3544560","title":"GPT-Based Automated Induction: Vulnerability Detection in Medical Software","year":2025,"lang":"en","type":"article","venue":"IEEE Journal of Biomedical and Health Informatics","topic":"Software Reliability and Analysis Research","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Brandon University","funders":"Ningbo Municipal Bureau of Science and Technology","keywords":"Computer science; Software; Vulnerability (computing); Computer security; Operating system","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007329052,0.0007060078,0.0005174934,0.001214797,0.0003607059,0.0006467834,0.00112611,0.0006501262,0.002542153],"category_scores_gemma":[0.004609743,0.0003826766,0.001119538,0.0006847986,0.001059475,0.001357581,0.001288299,0.001079494,0.0008180599],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008340071,"about_ca_system_score_gemma":0.001586707,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004096523,"about_ca_topic_score_gemma":0.004329006,"domain_scores_codex":[0.9992995,0.0001946266,0.00003771318,0.0001690266,0.0002226717,0.00007635984],"domain_scores_gemma":[0.9978708,0.001486863,0.0001631909,0.0002137182,0.000220526,0.00004486095],"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":[0.0002657366,0.0001469844,0.005413911,0.0003633613,0.00006746162,0.0008744302,0.0004454446,0.4632569,0.02611877,0.02800404,0.003839554,0.4712034],"study_design_scores_gemma":[0.000008263329,0.00002746398,0.0002094565,0.000009092114,0.00001098417,0.00009435546,0.00002060534,0.9802158,0.005552487,0.01320855,0.0006362149,0.000006655982],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02112838,0.0000708615,0.9737681,0.0001492323,0.00001366687,0.00005424333,0.0001307372,0.004080644,0.0006041738],"genre_scores_gemma":[0.6112263,0.0001636506,0.3854786,0.0001631299,0.0000238519,0.0001338927,0.0007047301,0.000531552,0.001574268],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004096523,"threshold_uncertainty_score":0.008504391,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02909851817937087,"score_gpt":0.3674956556795596,"score_spread":0.3383971375001887,"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."}}