{"id":"W4285490440","doi":"10.1145/3533767.3534220","title":"DocTer: documentation-guided fuzzing for testing deep learning API functions","year":2022,"lang":"en","type":"preprint","venue":"","topic":"Software Engineering Research","field":"Computer Science","cited_by":73,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"National Science Foundation","keywords":"Fuzz testing; Computer science; Documentation; Parsing; Function (biology); Constraint (computer-aided design); Software bug; Artificial intelligence; Dependency (UML); Software; Programming language; Machine learning; Data mining","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":[0.0009421071,0.0002615146,0.0002446006,0.0004110625,0.0005759765,0.0008527689,0.001407247,0.0001064951,0.0003252804],"category_scores_gemma":[0.003211668,0.0002923972,0.0001488757,0.0006282708,0.00001984175,0.0003554762,0.003201813,0.001087611,0.00005250274],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000384436,"about_ca_system_score_gemma":0.0002195911,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001887956,"about_ca_topic_score_gemma":0.000005798552,"domain_scores_codex":[0.997519,0.00009889125,0.0004005295,0.0008761785,0.0005812528,0.000524212],"domain_scores_gemma":[0.9952262,0.003449731,0.0001465034,0.0007716946,0.0002811307,0.0001247125],"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":[0.000004766448,0.00005231889,0.01843243,0.0004206756,0.000145916,0.00001843631,0.001338569,0.8769594,0.0005164408,0.00422931,0.004306778,0.09357497],"study_design_scores_gemma":[0.0006333149,0.0001802191,0.007499842,0.0001071063,0.00003255216,0.00003330273,0.000285399,0.9602835,0.000306382,0.004051087,0.02564828,0.0009390432],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.006483162,0.0001527096,0.9866881,0.0003160516,0.001744003,0.0007147542,0.000004132238,0.001831208,0.002065836],"genre_scores_gemma":[0.1898172,0.00001350054,0.797283,0.000122777,0.0004937422,0.002060848,0.0001443472,0.00008417736,0.009980417],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.1894052,"threshold_uncertainty_score":0.9999528,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06039005306412008,"score_gpt":0.331886397911752,"score_spread":0.2714963448476319,"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."}}