{"id":"W4403536220","doi":"10.1145/3691620.3695527","title":"Test-Driven Development and LLM-based Code Generation","year":2024,"lang":"en","type":"article","venue":"","topic":"Software Testing and Debugging Techniques","field":"Computer Science","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Test (biology); Code generation; Test-driven development; Programming language; Code (set theory); Software engineering; Software development; Operating system; Software","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":[0.0001364357,0.00006234019,0.00004586298,0.00006984481,0.00006577668,0.0002466395,0.0001379579,0.00002511818,0.000004071698],"category_scores_gemma":[0.00007601859,0.00005065993,0.000009705713,0.0001398164,0.00001223987,0.000112127,0.00006136049,0.00004231157,0.00002142486],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002329688,"about_ca_system_score_gemma":0.00009761425,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000009088826,"about_ca_topic_score_gemma":0.000007332614,"domain_scores_codex":[0.999494,0.000009684688,0.00008746199,0.0002193129,0.00009829765,0.00009119733],"domain_scores_gemma":[0.9995869,0.0001931331,0.000009140372,0.0001508675,0.00002402559,0.00003587125],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[6.612354e-7,0.00009069339,0.007297809,0.00007528549,0.00001287907,0.00006913231,0.0009501671,0.0001352778,0.003253781,0.02591408,0.1698128,0.7923874],"study_design_scores_gemma":[0.00003882013,0.00004437429,0.0006208868,0.00004278938,0.000001513597,0.0000115348,3.932532e-7,0.9621248,0.0220741,0.002591305,0.0123369,0.0001126458],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.007499058,0.0001262103,0.9794038,0.0007669297,0.00009919891,0.00005056273,3.890752e-7,0.01147042,0.0005834516],"genre_scores_gemma":[0.3946572,9.986958e-7,0.6048065,0.0003049861,0.00002333598,0.00001082238,0.000001526264,0.000003490231,0.0001911302],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9619895,"threshold_uncertainty_score":0.237835,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04352896262435825,"score_gpt":0.2753428871041018,"score_spread":0.2318139244797435,"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."}}