{"id":"W4396214523","doi":"10.1145/3649850","title":"PyDex: Repairing Bugs in Introductory Python Assignments using LLMs","year":2024,"lang":"en","type":"article","venue":"Proceedings of the ACM on Programming Languages","topic":"Software Testing and Debugging Techniques","field":"Computer Science","cited_by":31,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Python (programming language); Programming language; Computer science; Software engineering","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":[0.001852713,0.001446867,0.0006019085,0.001121522,0.0005513198,0.0009203794,0.00275021,0.001063537,0.008458891],"category_scores_gemma":[0.01179662,0.0007182306,0.0005741901,0.000617118,0.000977674,0.003448718,0.002885299,0.001554242,0.004589417],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007757277,"about_ca_system_score_gemma":0.001282562,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002523351,"about_ca_topic_score_gemma":0.003913584,"domain_scores_codex":[0.9983706,0.000362807,0.000143699,0.0005245254,0.000470522,0.0001279407],"domain_scores_gemma":[0.99505,0.002509062,0.0005688918,0.001083319,0.0005113723,0.0002773869],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001999063,0.0009261215,0.02871657,0.00193274,0.0001367783,0.001561125,0.004290213,0.01821535,0.04273649,0.003203007,0.1241806,0.7721019],"study_design_scores_gemma":[0.0006197235,0.002208439,0.03091272,0.0004384844,0.0002159007,0.002576987,0.002658491,0.5059935,0.2667195,0.01386005,0.1734272,0.0003690701],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"software","genre_gemma":"empirical","genre_scores_codex":[0.1690997,0.0004551554,0.1480205,0.0005756186,0.000172341,0.0003495486,0.00407169,0.6727226,0.00453279],"genre_scores_gemma":[0.6615663,0.0003139662,0.2940616,0.0007557975,0.00004366247,0.0003569915,0.01193225,0.01808441,0.01288495],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008458891,"threshold_uncertainty_score":0.02829778,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02296471450013041,"score_gpt":0.3016022634851895,"score_spread":0.2786375489850591,"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."}}