{"id":"W4407580437","doi":"10.1145/3717061","title":"An Empirical Study of Retrieval-Augmented Code Generation: Challenges and Opportunities","year":2025,"lang":"en","type":"article","venue":"ACM Transactions on Software Engineering and Methodology","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":20,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University","funders":"","keywords":"Computer science; Code (set theory); Data science; Information retrieval; Programming language","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":[0.02147613,0.001058443,0.0006090318,0.002125335,0.000996602,0.002229719,0.001887005,0.001462716,0.002555077],"category_scores_gemma":[0.1739769,0.0006746629,0.000833708,0.002285217,0.001963749,0.008249703,0.002432816,0.003075063,0.001268838],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001069858,"about_ca_system_score_gemma":0.001455701,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003689805,"about_ca_topic_score_gemma":0.004335057,"domain_scores_codex":[0.9808797,0.01014311,0.001340928,0.002656957,0.004516495,0.0004627679],"domain_scores_gemma":[0.785157,0.1700215,0.01021373,0.01941073,0.0134252,0.00177184],"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":[0.002260276,0.002491333,0.4251079,0.003732095,0.0005470207,0.001123992,0.008799589,0.03004484,0.009897376,0.007514881,0.02992001,0.4785607],"study_design_scores_gemma":[0.0007413123,0.006240651,0.4096999,0.001931084,0.0007746839,0.004658395,0.01485317,0.4106149,0.03185183,0.01888646,0.09926211,0.0004855858],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9565821,0.003781275,0.02792201,0.001347113,0.000153639,0.0004259657,0.002172883,0.001608614,0.006006331],"genre_scores_gemma":[0.968127,0.0008456453,0.02253157,0.0003685262,0.00005725451,0.0003491556,0.00572072,0.0004035268,0.001596711],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02147613,"threshold_uncertainty_score":0.113578,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.209148202889168,"score_gpt":0.3915554796441723,"score_spread":0.1824072767550042,"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."}}