{"id":"W7123348570","doi":"10.1109/esem64174.2025.00021","title":"Secret Breach Detection in Source Code with Large Language Models","year":2025,"lang":"","type":"article","venue":"","topic":"Advanced Malware Detection Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"False positive paradox; Source code; Benchmark (surveying); Binary number; Scalability; Code (set theory); Binary code; Key (lock)","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":[0.0005578035,0.0003803148,0.0003833525,0.0007317254,0.0002285087,0.000220715,0.000771546,0.0002581006,0.00005262308],"category_scores_gemma":[0.00004276939,0.0003635961,0.00008170561,0.002309257,0.00008499398,0.001335117,0.0005193747,0.0006586586,0.00001735292],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003459883,"about_ca_system_score_gemma":0.0001676337,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006353732,"about_ca_topic_score_gemma":0.005895225,"domain_scores_codex":[0.9972298,0.0001779658,0.0005057115,0.001019067,0.0003675246,0.000699945],"domain_scores_gemma":[0.9983981,0.0001063411,0.000161129,0.001061034,0.0001664352,0.0001069707],"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":[0.000329701,0.0005474337,0.0003653552,0.0002251769,0.00007585054,0.0001186719,0.006681833,0.02248753,0.009664151,0.03264285,0.0002532048,0.9266083],"study_design_scores_gemma":[0.0009825272,0.000268096,0.0001606175,0.0001979814,0.00001402117,0.00004489911,0.0007597603,0.8040022,0.183527,0.006793165,0.002818099,0.000431655],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01451521,0.0003116526,0.971829,0.0003029422,0.0002270587,0.0006682726,0.000009032407,0.001274621,0.0108622],"genre_scores_gemma":[0.9439512,0.0000463752,0.04604964,0.0007948221,0.00003180459,0.0001040403,0.0000020959,0.00002886521,0.008991116],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.929436,"threshold_uncertainty_score":0.9998816,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006949916592829725,"score_gpt":0.2561849373312464,"score_spread":0.2492350207384166,"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."}}