{"id":"W4401632577","doi":"10.22215/etd/2024-16009","title":"Multilingual Fault Localization for Deep Learning Compilers","year":2024,"lang":"en","type":"dissertation","venue":"","topic":"Software Testing and Debugging Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University","funders":"","keywords":"Compiler; Codebase; Computer science; Deep learning; Programming language; Artificial intelligence; 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":[0.0002542014,0.0002390021,0.0002229888,0.0002389926,0.0001818454,0.0003260092,0.000524717,0.0002535198,0.000007518285],"category_scores_gemma":[0.0005624655,0.0002252534,0.0001264586,0.0002957126,0.00001251463,0.0001199809,0.00005452419,0.0003073943,0.00004302232],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000544224,"about_ca_system_score_gemma":0.00008747028,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007373178,"about_ca_topic_score_gemma":0.00007079069,"domain_scores_codex":[0.9986482,0.00003092702,0.0002831566,0.0005612256,0.0002373988,0.0002391075],"domain_scores_gemma":[0.9989023,0.0003740333,0.0001245471,0.0002871252,0.000256919,0.00005508851],"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":[0.00002672763,0.00006286852,0.0002029427,0.001080529,0.00007853968,0.00001844931,0.008662705,0.004989113,0.0000660511,0.009501232,0.04990277,0.9254081],"study_design_scores_gemma":[0.0001060296,0.0001021743,0.00003310511,0.000313286,0.00002749991,0.000004632707,0.0001114079,0.9652055,0.001760381,0.02168567,0.01029643,0.0003538855],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.000749101,0.0004399425,0.9700778,0.00003299382,0.001374168,0.0003356884,0.000001383392,0.02355614,0.003432741],"genre_scores_gemma":[0.2518575,0.00005204065,0.7152812,0.0003638885,0.0004060939,0.0003199843,0.001861376,0.0001586784,0.02969918],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9602164,"threshold_uncertainty_score":0.9185568,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01955025092504285,"score_gpt":0.3159425092879183,"score_spread":0.2963922583628755,"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."}}