{"id":"W7123350572","doi":"10.1109/esem64174.2025.00036","title":"Is LLM-Generated Code More Maintainable & Reliable Than Human-Written Code?","year":2025,"lang":"","type":"article","venue":"","topic":"Software Engineering Research","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University","funders":"","keywords":"Software quality; Maintainability; Python (programming language); Code review; Code (set theory); Software; Quality (philosophy); Key (lock)","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":[0.01210805,0.0004593376,0.000394178,0.001566821,0.0003987403,0.00274059,0.001379387,0.0009441182,0.001823263],"category_scores_gemma":[0.1286713,0.0003660017,0.0004997282,0.0009605999,0.00180299,0.002381762,0.001714314,0.0009290808,0.0007101233],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001122154,"about_ca_system_score_gemma":0.001662246,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002369253,"about_ca_topic_score_gemma":0.003824229,"domain_scores_codex":[0.9883117,0.005377899,0.000599361,0.001957495,0.003425877,0.0003277165],"domain_scores_gemma":[0.8795105,0.07580779,0.01443164,0.01752121,0.01105661,0.001672251],"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":[0.001293112,0.0006167798,0.3077122,0.003653328,0.0004575059,0.0008332927,0.01105976,0.04270103,0.04291132,0.005225784,0.0158632,0.5676727],"study_design_scores_gemma":[0.0003315273,0.003047491,0.381717,0.002277364,0.0004674387,0.002303872,0.00779801,0.4286526,0.06920277,0.02579368,0.07797721,0.0004311339],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8994312,0.001068091,0.08320504,0.001862069,0.0001286509,0.0002207682,0.00135895,0.006286695,0.006438496],"genre_scores_gemma":[0.9379966,0.0002212568,0.05741646,0.0003667982,0.00003214581,0.0001225733,0.001877767,0.0009203791,0.001046072],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01210805,"threshold_uncertainty_score":0.06403428,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03301275349302725,"score_gpt":0.3218060359485615,"score_spread":0.2887932824555343,"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."}}