{"id":"W4414682256","doi":"10.48550/arxiv.2506.19045","title":"Efficient Black-Box Fault Localization for System-Level Test Code Using Large Language Models","year":2025,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Software Testing and Debugging Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Code (set theory); TRACE (psycholinguistics); Pruning; Test suite; Code coverage; Inference; Source code; Test case; Debugging; Fault (geology)","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004741585,0.0003424079,0.0003829818,0.0003918149,0.0002884479,0.0001509828,0.001445839,0.0003369049,0.000001131179],"category_scores_gemma":[0.0002492472,0.0003978708,0.0001978473,0.0006687545,0.00006512497,0.000119222,0.001564385,0.0002940982,0.000005973763],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000507633,"about_ca_system_score_gemma":0.0003194955,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003124581,"about_ca_topic_score_gemma":0.00002293659,"domain_scores_codex":[0.9979686,0.00009647843,0.0002708995,0.00110431,0.0001269412,0.0004327359],"domain_scores_gemma":[0.9974643,0.0004815125,0.0002958166,0.001258434,0.0003898627,0.0001100792],"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":[0.000007543538,0.0001094369,0.0003653024,0.0004619253,0.0000270273,0.00005136253,0.0004452278,0.9397267,0.000009375094,0.0579326,0.0007528588,0.0001106881],"study_design_scores_gemma":[0.0003620257,0.0000306346,0.00001157787,0.0008227477,0.00008071998,0.000004191966,0.00007171767,0.9838405,0.0001720339,0.01417133,0.00003964793,0.0003928848],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01361379,0.00005723275,0.9782396,0.00001763763,0.0003616135,0.0006583172,0.0002982197,0.006196684,0.0005569294],"genre_scores_gemma":[0.9245339,0.00000586772,0.07460333,0.00008986721,0.00005064664,0.000003726961,0.00004203899,0.00002147665,0.000649162],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9109201,"threshold_uncertainty_score":0.9998473,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1009006671050115,"score_gpt":0.2378397593333464,"score_spread":0.1369390922283349,"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."}}