{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009984892,0.0001906952,0.000198962,0.0002843182,0.0000901704,0.0003195602,0.002136603,0.00007791031,0.000001474818],"category_scores_gemma":[0.002545661,0.0001399059,0.000100439,0.0008912398,0.00007576527,0.0003845297,0.001200882,0.0003267503,0.000003074438],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000133891,"about_ca_system_score_gemma":0.00004115289,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001092006,"about_ca_topic_score_gemma":0.000001090763,"domain_scores_codex":[0.9983932,0.00001647893,0.0002986154,0.0005265693,0.0004107401,0.0003544091],"domain_scores_gemma":[0.9988913,0.0001950701,0.0001366878,0.0006550771,0.00007760933,0.00004424143],"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":[0.00003270324,0.0003759868,0.09573543,0.001441411,0.0001067567,0.00006893549,0.01133956,0.00006485149,0.06088803,0.01611809,0.01237003,0.8014582],"study_design_scores_gemma":[0.001337494,0.00144285,0.02591796,0.02127973,0.0001950276,0.0006388712,0.001567816,0.06679298,0.6209052,0.24288,0.01398389,0.003058194],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9832026,0.001506299,0.001855845,0.001805309,0.0006185505,0.0005108527,0.000001109235,0.009931546,0.0005678689],"genre_scores_gemma":[0.7647365,0.000005270509,0.2349369,0.00007452993,0.0001080095,0.00002618257,2.463254e-7,0.00002571107,0.00008661055],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7984,"threshold_uncertainty_score":0.5705196,"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."}}