{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001368687,0.002710229,0.001227538,0.002644083,0.0005596941,0.00132188,0.002478188,0.001479846,0.00185807],"category_scores_gemma":[0.01137005,0.0008158361,0.002005055,0.001019467,0.0008591319,0.002173207,0.001649456,0.0022355,0.002500746],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001251241,"about_ca_system_score_gemma":0.002234719,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009287605,"about_ca_topic_score_gemma":0.0180164,"domain_scores_codex":[0.9981934,0.0004324486,0.0001204003,0.0006202213,0.000484313,0.0001491761],"domain_scores_gemma":[0.9918963,0.004957317,0.0009780211,0.001074571,0.0008785509,0.0002152619],"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":[0.0007853716,0.0006586273,0.03137495,0.00103143,0.0002568317,0.001037741,0.0007374894,0.2972985,0.04156391,0.003806376,0.01883775,0.6026109],"study_design_scores_gemma":[0.00002972523,0.00009513796,0.001022166,0.00003293912,0.00002644797,0.0001571096,0.00008038006,0.985104,0.008844126,0.003016755,0.001573413,0.00001779477],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1215538,0.001353726,0.776191,0.000631012,0.0000918345,0.0002487736,0.002413957,0.09643012,0.001085713],"genre_scores_gemma":[0.566228,0.000358036,0.4158059,0.00045712,0.0000481664,0.0003684443,0.0113853,0.003174955,0.002174014],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.009287605,"threshold_uncertainty_score":0.01846707,"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."}}