{"id":"W4417436230","doi":"10.1145/3785363","title":"Galápagos: Automated N-Version Programming with LLMs","year":2025,"lang":"en","type":"article","venue":"ACM Transactions on Software Engineering and Methodology","topic":"Software Testing and Debugging Techniques","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Correctness; Redundancy (engineering); Software; Symbolic execution; Program analysis; Automatic programming; Software quality; Programming paradigm; Software testing","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":[0.0005209685,0.0001889249,0.0002353498,0.0003707302,0.0001789414,0.00006139717,0.0004086487,0.0001333394,0.000002277636],"category_scores_gemma":[0.0008958234,0.0001661989,0.00004473586,0.0006051086,0.0000458134,0.0001341091,0.00002711689,0.0002805028,0.000002793535],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003768907,"about_ca_system_score_gemma":0.00005054304,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000419957,"about_ca_topic_score_gemma":0.000001626455,"domain_scores_codex":[0.9989311,0.0001146751,0.0001621129,0.0004028821,0.0001060634,0.0002832052],"domain_scores_gemma":[0.9971249,0.0020955,0.00003525863,0.0006052578,0.00006642104,0.00007265815],"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":[0.00006333141,0.0001249106,0.0009925452,0.0001973755,0.0001625605,0.00006015909,0.000600577,0.01565865,0.0003322932,0.001703902,0.0009269103,0.9791768],"study_design_scores_gemma":[0.006878985,0.005039911,0.01581764,0.003913817,0.000645723,0.002460371,0.0001916571,0.8121606,0.05172582,0.0275713,0.06923115,0.00436304],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.003194809,0.000132785,0.9706535,0.0002928461,0.0002955776,0.0001334927,0.000001596274,0.02528453,0.00001080877],"genre_scores_gemma":[0.06623493,0.00002082536,0.9333974,0.000165222,0.000007835572,0.00005978187,0.000001533381,0.00001415531,0.00009831004],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9748138,"threshold_uncertainty_score":0.6777393,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04061881690512797,"score_gpt":0.3080365130385907,"score_spread":0.2674176961334627,"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."}}