{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005121061,0.001026674,0.0005169607,0.0009871738,0.000526232,0.0009475536,0.001719511,0.001714379,0.001569238],"category_scores_gemma":[0.03573749,0.0005928908,0.000644773,0.0005223839,0.001446016,0.002413653,0.001383772,0.002540188,0.0005392524],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00135165,"about_ca_system_score_gemma":0.001220934,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008922956,"about_ca_topic_score_gemma":0.01482413,"domain_scores_codex":[0.9971963,0.001716107,0.00009787315,0.0005324847,0.0003317549,0.0001255063],"domain_scores_gemma":[0.9561964,0.03912226,0.00100659,0.00204218,0.001285277,0.0003471498],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0006690684,0.0008243561,0.02155022,0.000676582,0.0002057136,0.0008983324,0.002166665,0.8499894,0.008743327,0.004981066,0.008630355,0.1006649],"study_design_scores_gemma":[0.00005899735,0.0002321429,0.001706123,0.00003171348,0.00002686867,0.0001408231,0.0003684779,0.9874246,0.004125804,0.004283469,0.001576205,0.00002491643],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8943498,0.0009576162,0.09484602,0.001384863,0.0001146449,0.0001686646,0.0008431315,0.004984567,0.002350893],"genre_scores_gemma":[0.9575008,0.0001400233,0.03909775,0.0002658395,0.00002463534,0.0001211778,0.001444122,0.000346886,0.001058828],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.008922956,"threshold_uncertainty_score":0.0270831,"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."}}