{"id":"W4413308446","doi":"10.1007/s10664-025-10707-0","title":"Towards understanding the challenges of bug localization in deep learning systems","year":2025,"lang":"en","type":"article","venue":"Empirical Software Engineering","topic":"Software Engineering Research","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":false,"ca_institutions":"Dalhousie University","funders":"Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Deep learning; Data science; Artificial intelligence; Software engineering","routes":{"ca_aff":true,"ca_fund":true,"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":[0.00071965,0.0001678917,0.0002478986,0.0003871741,0.00007483425,0.00008829025,0.0007211612,0.000113555,0.000002018897],"category_scores_gemma":[0.003007012,0.0001407465,0.00005933852,0.001389732,0.00003050839,0.0001950677,0.000299999,0.0004314068,0.000003397436],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000394116,"about_ca_system_score_gemma":0.00006757546,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003254796,"about_ca_topic_score_gemma":0.000004375425,"domain_scores_codex":[0.9984871,0.00008308285,0.0003357603,0.0003172896,0.0003915507,0.0003852386],"domain_scores_gemma":[0.9976309,0.001816022,0.00004375242,0.0003833251,0.00006550797,0.00006047807],"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":[0.000002561167,0.00002128253,0.0622778,0.0002795303,0.00002645511,0.000009917972,0.001054961,0.910623,0.00001566836,0.02332962,0.00004863977,0.002310577],"study_design_scores_gemma":[0.0002996636,0.00004705014,0.07574259,0.0004469781,0.000005448735,0.000006373176,0.0002715913,0.9204323,0.0001402155,0.0008431447,0.001531318,0.0002333466],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006976232,0.004434514,0.9868532,0.0004787993,0.0004861508,0.000182084,2.935219e-7,0.0004963758,0.00009229438],"genre_scores_gemma":[0.9967422,0.0001043445,0.003018277,0.00002269428,0.00003611356,0.0000361399,7.925364e-7,0.00001811265,0.00002134989],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9897659,"threshold_uncertainty_score":0.5739475,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04201334118444038,"score_gpt":0.2910416336842223,"score_spread":0.2490282924997819,"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."}}