{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003997441,0.00006682135,0.00008549174,0.00009048517,0.0001766613,0.00006203757,0.0004463575,0.0000239699,0.00001032272],"category_scores_gemma":[0.005253917,0.00004600687,0.0000324816,0.00106904,0.00003070137,0.00007770637,0.0001344493,0.0001111508,0.00002861193],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001968231,"about_ca_system_score_gemma":0.00005228764,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000441105,"about_ca_topic_score_gemma":7.544887e-7,"domain_scores_codex":[0.9987597,0.0002449299,0.0001830487,0.000177325,0.0004836475,0.0001513725],"domain_scores_gemma":[0.9969468,0.002145089,0.0001230707,0.0003518727,0.0004132079,0.00001999131],"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":[5.861473e-7,0.000005572636,0.03161388,0.00001845348,0.000004122974,0.000001209103,0.0002088429,0.0424525,0.0001544795,0.0009307925,0.001036499,0.9235731],"study_design_scores_gemma":[0.00007756743,0.00002214937,0.03058426,0.00005745418,0.000006641688,0.000002096974,0.000008270061,0.9609895,0.000502157,0.007624763,0.00006575848,0.00005934532],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02810323,0.0000678324,0.9480744,0.001156918,0.00008971099,0.0002037,1.544513e-7,0.01688406,0.005420012],"genre_scores_gemma":[0.8944204,3.477765e-7,0.1053743,0.0001142912,0.00004383639,0.00001693193,0.000001601362,0.00000579722,0.00002252431],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9235137,"threshold_uncertainty_score":0.6289809,"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."}}