{"id":"W4411271588","doi":"10.1109/msr66628.2025.00038","title":"Combining Large Language Models with Static Analyzers for Code Review Generation","year":2025,"lang":"en","type":"article","venue":"","topic":"Software Engineering Research","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Computer science; Programming language; Code generation; Code (set theory); Key (lock); Operating system","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":[0.005362904,0.00177351,0.001135034,0.00486951,0.0006288547,0.002217684,0.002680098,0.001386666,0.003114055],"category_scores_gemma":[0.03781271,0.0008241443,0.001500642,0.002085264,0.0006452654,0.003491896,0.002036555,0.002818109,0.004962697],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00148808,"about_ca_system_score_gemma":0.003735552,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006656746,"about_ca_topic_score_gemma":0.0191902,"domain_scores_codex":[0.9943854,0.002249931,0.0005348115,0.001505026,0.001129142,0.0001958011],"domain_scores_gemma":[0.9692617,0.01926251,0.002059332,0.003703473,0.005176928,0.0005360944],"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":[0.0004005468,0.0004526866,0.01384683,0.0007834264,0.0002616029,0.0004541217,0.0006157353,0.05740172,0.01743399,0.003561355,0.03387903,0.870909],"study_design_scores_gemma":[0.0001004183,0.0001632334,0.001785201,0.00008445129,0.0001227701,0.0002062452,0.0001329642,0.9578151,0.01585825,0.01126159,0.01240817,0.0000616317],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0398754,0.001508668,0.8613852,0.001192746,0.0003051514,0.0007205338,0.004005616,0.08852391,0.002482602],"genre_scores_gemma":[0.2862033,0.0005730269,0.690415,0.0009771166,0.0002316976,0.0008865186,0.01533287,0.002470905,0.002909667],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006656746,"threshold_uncertainty_score":0.02836204,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03049753377986934,"score_gpt":0.324670604131976,"score_spread":0.2941730703521067,"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."}}