{"id":"W3196351123","doi":"","title":"Leveraging Documentation to Test Deep Learning Library Functions","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Software Testing and Debugging Techniques","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Fuzz testing; Constraint (computer-aided design); Code (set theory); Artificial intelligence; Software bug; Machine learning; Data mining; Software; Programming language; Set (abstract data type)","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0001526022,0.0002443237,0.0002134584,0.0003548358,0.0002962026,0.0006281957,0.001020317,0.0001543963,0.00006088403],"category_scores_gemma":[0.0002581248,0.0003157417,0.0001313383,0.000947573,0.00002402558,0.0008809923,0.002465255,0.0006184831,0.00008571937],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001352825,"about_ca_system_score_gemma":0.0001568399,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001433494,"about_ca_topic_score_gemma":0.000007435996,"domain_scores_codex":[0.9983162,0.000128235,0.000165893,0.00100828,0.00009342428,0.0002879281],"domain_scores_gemma":[0.9983345,0.0004163343,0.000147971,0.0008240048,0.0001072578,0.0001699352],"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":[0.00001601241,0.0003319074,0.1693174,0.000150133,0.0001266469,0.001175247,0.003430276,0.7804075,0.0001269552,0.007421239,0.0094085,0.02808822],"study_design_scores_gemma":[0.0003747317,0.0002289959,0.009485544,0.0006950084,0.00009595306,0.00003274066,0.000336779,0.9396623,0.0007015917,0.04500953,0.002081725,0.001295086],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1052964,0.00005126144,0.8828303,0.0002127643,0.0003935242,0.0001569658,0.000001516265,0.009276674,0.001780634],"genre_scores_gemma":[0.9378244,0.00004021281,0.05966212,0.0002732635,0.00007107615,0.000002838132,0.00004164722,0.00002412009,0.002060376],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.832528,"threshold_uncertainty_score":0.9999295,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05306870260441219,"score_gpt":0.1940805870896694,"score_spread":0.1410118844852572,"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."}}