{"id":"W4385187421","doi":"10.1109/sp46215.2023.10179420","title":"Examining Zero-Shot Vulnerability Repair with Large Language Models","year":2023,"lang":"en","type":"article","venue":"","topic":"Software Testing and Debugging Techniques","field":"Computer Science","cited_by":178,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"","keywords":"Zero (linguistics); Computer science; Vulnerability (computing); Ground zero; Computer security; Physics; Linguistics; Nuclear physics","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":[0.0009948318,0.0001296098,0.000143628,0.0001156917,0.0001391435,0.00008586352,0.0004978678,0.00004947151,0.00001165176],"category_scores_gemma":[0.0002059448,0.00009701772,0.00003848615,0.0007605345,0.0000283925,0.00035713,0.000327572,0.0001368062,0.0000725284],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002494656,"about_ca_system_score_gemma":0.00004374569,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009363167,"about_ca_topic_score_gemma":0.00001274869,"domain_scores_codex":[0.998686,0.00006724424,0.0001524341,0.000471103,0.0002635185,0.0003596862],"domain_scores_gemma":[0.9984936,0.0003756176,0.00003979551,0.0009539632,0.0000590954,0.00007792358],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004721493,0.000579318,0.1628423,0.0002489634,0.0001389828,0.001235702,0.02600465,0.005154631,0.0009278814,0.1769191,0.5435123,0.08238888],"study_design_scores_gemma":[0.0002963881,0.0002355253,0.006326729,0.00006950757,0.000005681428,0.00003877035,0.00007172536,0.9233274,0.001046405,0.06776404,0.0004249555,0.0003928248],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1048523,0.0000192845,0.8097978,0.0001183022,0.00005523905,0.0001050803,0.000001715579,0.07664462,0.008405714],"genre_scores_gemma":[0.7341148,0.000001052392,0.26456,0.0002745136,0.0000171299,0.0000254457,0.000003347113,0.00001105306,0.0009926218],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9181728,"threshold_uncertainty_score":0.3956268,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07209621342546572,"score_gpt":0.3021107687630803,"score_spread":0.2300145553376146,"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."}}