{"id":"W4406460194","doi":"10.1109/tps-isa62245.2024.00043","title":"Fine-Tuning LLMs for Code Mutation: A New Era of Cyber Threats","year":2024,"lang":"en","type":"article","venue":"","topic":"Software Testing and Debugging Techniques","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ontario Tech University","funders":"","keywords":"Computer science; Computer security; Code (set theory); Mutation; Internet privacy; Programming language; Genetics; Biology","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":[],"consensus_categories":[],"category_scores_codex":[0.0001744875,0.00007349696,0.00009533517,0.00005897409,0.00003689888,0.0001104235,0.0002848353,0.00003340791,0.0000137145],"category_scores_gemma":[0.0001250656,0.00006002407,0.00005553323,0.0002664471,0.00001463734,0.0001817327,0.00007124099,0.00006717637,0.00001064115],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001317487,"about_ca_system_score_gemma":0.0001140307,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009743233,"about_ca_topic_score_gemma":0.00001582712,"domain_scores_codex":[0.9993929,0.00000852632,0.0001422551,0.0002235549,0.0001110629,0.0001217062],"domain_scores_gemma":[0.999095,0.0005162454,0.00002574456,0.0002634996,0.00005767094,0.00004183288],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.000002978349,0.00001761522,0.0003106295,0.00007116175,0.00002079914,0.00002382015,0.001153501,0.00007786956,0.0002088226,0.107736,0.1992144,0.6911624],"study_design_scores_gemma":[0.0001738158,0.0001781849,0.0002553787,0.0003511266,0.00001221731,0.00008028886,0.000003616826,0.3972148,0.008264395,0.5891808,0.004092573,0.0001928047],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0007799341,0.0004526571,0.98716,0.001733449,0.0002151144,0.0001082973,0.000001797843,0.007919665,0.00162912],"genre_scores_gemma":[0.2307115,0.000002066247,0.7671143,0.0001431222,0.00006366521,0.00001498219,0.000002117382,0.000008313869,0.001939924],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.6909696,"threshold_uncertainty_score":0.2447711,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04922846117780114,"score_gpt":0.323486207207921,"score_spread":0.2742577460301198,"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."}}