{"id":"W4411523084","doi":"10.1145/3728947","title":"The First Prompt Counts the Most! An Evaluation of Large Language Models on Iterative Example-Based Code Generation","year":2025,"lang":"en","type":"article","venue":"Proceedings of the ACM on software engineering.","topic":"Software Engineering Research","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"","keywords":"Code (set theory); Computer science; Benchmark (surveying); Iterative and incremental development; Code generation; Process (computing); Natural language generation; Natural language; Programming language; Software engineering; Artificial intelligence; Computer security; Geography","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":[0.001670828,0.0001495477,0.0001265205,0.000114238,0.0001749838,0.0001180673,0.002068869,0.00005969264,0.000003278227],"category_scores_gemma":[0.005430353,0.0000874509,0.00005477722,0.0006053236,0.00003533925,0.0002025471,0.0002941588,0.0002251554,0.000002976113],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001213845,"about_ca_system_score_gemma":0.0001039861,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00000936921,"about_ca_topic_score_gemma":0.000003904341,"domain_scores_codex":[0.9984741,0.00002145925,0.0002088227,0.0002558232,0.0008197862,0.0002199874],"domain_scores_gemma":[0.9976984,0.0007950236,0.00009421659,0.0008240567,0.0005600543,0.0000282925],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001280448,0.0004833721,0.005905446,0.0005372062,0.0002585542,9.450579e-7,0.006495096,0.7683342,0.00667116,0.1584479,0.03070466,0.02203338],"study_design_scores_gemma":[0.0003684688,0.00009088223,0.004046275,0.000159626,0.00001673583,6.674866e-7,0.0000234779,0.9609544,0.03222112,0.0006862665,0.001317196,0.0001148327],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8017045,0.0006660337,0.1907818,0.002473331,0.001174522,0.002139442,0.00005223846,0.0006200733,0.000388083],"genre_scores_gemma":[0.9931406,0.000004119664,0.006373598,0.0001451951,0.00006024957,0.0001540499,0.000003619267,0.00001916941,0.00009933393],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1926202,"threshold_uncertainty_score":0.6501032,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03728653452721785,"score_gpt":0.2970827131643091,"score_spread":0.2597961786370912,"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."}}