{"id":"W4385768129","doi":"10.24963/ijcai.2023/542","title":"Revisiting the Evaluation of Deep Learning-Based Compiler Testing","year":2023,"lang":"en","type":"article","venue":"","topic":"Software Testing and Debugging Techniques","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Compiler; Generator (circuit theory); Correctness; Programming language; Optimizing compiler; Domain-specific language; Compiler construction; Artificial intelligence","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005608286,0.002557912,0.00116282,0.002248969,0.0005478872,0.00184796,0.004640687,0.001954155,0.002770092],"category_scores_gemma":[0.03265042,0.0007948463,0.0012236,0.001654317,0.002079292,0.004753576,0.002552573,0.003376903,0.001536509],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002905496,"about_ca_system_score_gemma":0.00340466,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00876802,"about_ca_topic_score_gemma":0.01398246,"domain_scores_codex":[0.9893394,0.003592208,0.0009228078,0.002358564,0.003098114,0.0006889537],"domain_scores_gemma":[0.9824091,0.008622955,0.0009414209,0.003225881,0.004157573,0.0006429657],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.002182928,0.00136511,0.02348043,0.003031507,0.0006659208,0.0004361609,0.0003566967,0.2148397,0.02620991,0.006771423,0.05462285,0.6660374],"study_design_scores_gemma":[0.0004456695,0.001426993,0.00770755,0.0005186617,0.0001856916,0.0004077526,0.0002352746,0.8961121,0.05559921,0.009739805,0.02750306,0.0001182351],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6546206,0.01511281,0.203906,0.003930916,0.00233217,0.0008699616,0.008965599,0.08299173,0.02727024],"genre_scores_gemma":[0.7964147,0.001414418,0.1698185,0.002028057,0.0001446281,0.0003618064,0.02201355,0.002650848,0.005153399],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.00876802,"threshold_uncertainty_score":0.02965981,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1135955612960345,"score_gpt":0.3410612020110207,"score_spread":0.2274656407149863,"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."}}