{"id":"W7127344164","doi":"10.1109/ccece64018.2025.11364452","title":"How Effective are pretrained Programming Language-based Language Models (PLLMs) in the Detection of Android Vulnerabilities?","year":2025,"lang":"","type":"article","venue":"","topic":"Advanced Malware Detection Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Lethbridge","funders":"Science and Engineering Research Council","keywords":"Android (operating system); Language model; Vulnerability (computing); Mobile device; Multiple Models; Android application","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"],"consensus_categories":[],"category_scores_codex":[0.00147567,0.000429186,0.0005397176,0.0007256728,0.0002199553,0.0003838614,0.001060994,0.0002730649,0.000006818445],"category_scores_gemma":[0.000765269,0.0003436882,0.000239519,0.002647516,0.0002780753,0.00106674,0.0002174227,0.0007375705,7.498372e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003277806,"about_ca_system_score_gemma":0.0001393051,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004620745,"about_ca_topic_score_gemma":0.0009535227,"domain_scores_codex":[0.996508,0.0009443864,0.0005615292,0.0008843302,0.0005399228,0.0005618528],"domain_scores_gemma":[0.9970306,0.0009223694,0.0004145548,0.001271961,0.0003022777,0.00005827301],"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":[0.0001701857,0.0003715381,0.000110902,0.0007047122,0.00004199857,0.00003853813,0.0123697,0.002927145,0.02340217,0.002988399,0.00001126236,0.9568635],"study_design_scores_gemma":[0.001195362,0.0008119695,0.0005371852,0.0004924597,0.00003536754,0.00001097297,0.01499408,0.3479227,0.6297119,0.00376885,0.0001140903,0.0004051638],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06138283,0.0009309798,0.9322306,0.0007559589,0.0002635312,0.003166624,0.00001223329,0.0004970871,0.0007601368],"genre_scores_gemma":[0.9741169,0.00001056449,0.02409613,0.0003125258,0.00004284305,0.0009206127,0.000002406648,0.00002180427,0.0004762406],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9564583,"threshold_uncertainty_score":0.9999015,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008168571883936345,"score_gpt":0.2613502313963667,"score_spread":0.2531816595124304,"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."}}