{"id":"W4416961966","doi":"10.1109/pst65910.2025.11268838","title":"RefPentester: A Knowledge-Informed Self-Reflective Penetration Testing Framework Based on Large Language Models","year":2025,"lang":"","type":"article","venue":"","topic":"Software Testing and Debugging Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal; Concordia University","funders":"Concordia University","keywords":"Process (computing); Hacker; Baseline (sea); Fuzz testing; Soundness","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001306625,0.0006374395,0.0005253373,0.0008341769,0.0007609426,0.0008958263,0.001385522,0.0004563494,0.00003905998],"category_scores_gemma":[0.005132935,0.0006151836,0.000193342,0.003353127,0.00006620227,0.0008220147,0.0005508188,0.0009381364,0.000118648],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006321503,"about_ca_system_score_gemma":0.001834108,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000132941,"about_ca_topic_score_gemma":0.0000618088,"domain_scores_codex":[0.9960721,0.0003104576,0.0008227652,0.001240894,0.0005508732,0.001002932],"domain_scores_gemma":[0.9908266,0.006247894,0.0003234374,0.001694736,0.0007028205,0.0002045293],"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":[0.0003861537,0.008739623,0.02135093,0.002215478,0.0004519668,0.0002148865,0.04840821,0.01400832,0.0002503883,0.3597492,0.07177803,0.4724468],"study_design_scores_gemma":[0.0006198128,0.0006496985,0.0007415551,0.003084457,0.00005304049,0.000006196774,0.0000566275,0.9206434,0.001249433,0.07215202,0.0001826381,0.0005611284],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001290654,0.0003119764,0.8445764,0.0004802759,0.0006604838,0.0007707741,0.000006700868,0.01760454,0.1342982],"genre_scores_gemma":[0.5489393,0.000005447363,0.4478815,0.002196421,0.00008418264,0.00008395536,0.000005189528,0.00002380416,0.0007801367],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9066351,"threshold_uncertainty_score":0.99963,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03270531685287602,"score_gpt":0.3436117691507119,"score_spread":0.3109064522978359,"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."}}