{"id":"W4415745948","doi":"10.1109/icsme64153.2025.00083","title":"Prompting Matters: Assessing the Effect of Prompting Techniques on LLM-Generated Class Code","year":2025,"lang":"","type":"article","venue":"","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"","keywords":"Correctness; Software quality; Workflow; Code review; Coding (social sciences); Control flow; Class (philosophy); Consistency (knowledge bases); Static program analysis","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":[0.009686006,0.001320871,0.0007078815,0.001549847,0.0003704425,0.00139739,0.0009106012,0.00126748,0.001311182],"category_scores_gemma":[0.1457204,0.0003030627,0.0006670531,0.0007625547,0.0007180914,0.001589958,0.001211938,0.001370671,0.0008173512],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005741921,"about_ca_system_score_gemma":0.0009158651,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001105718,"about_ca_topic_score_gemma":0.001311416,"domain_scores_codex":[0.9896372,0.00488109,0.001170669,0.002020064,0.002001378,0.0002896127],"domain_scores_gemma":[0.7665678,0.2033008,0.01291507,0.007174589,0.007892887,0.002148815],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.01321607,0.004673724,0.1465841,0.004374247,0.0007325524,0.0007576479,0.01044156,0.02403691,0.09124999,0.0007202292,0.008030019,0.6951829],"study_design_scores_gemma":[0.002581891,0.05067882,0.5369186,0.00145277,0.00239759,0.002159981,0.008825682,0.1695824,0.1921445,0.004969877,0.02744942,0.0008384244],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9798194,0.0007218394,0.01266701,0.0002035543,0.0001027123,0.0004878916,0.0004508737,0.004096451,0.001450346],"genre_scores_gemma":[0.9554075,0.0002810611,0.04053312,0.0002027019,0.0000440724,0.0006410367,0.001310139,0.0005519468,0.00102837],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.009686006,"threshold_uncertainty_score":0.05122513,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1063125903891533,"score_gpt":0.4805392003762449,"score_spread":0.3742266099870917,"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."}}