{"id":"W4319594569","doi":"10.1145/3583566","title":"COMET: Coverage-guided Model Generation For Deep Learning Library Testing","year":2023,"lang":"en","type":"article","venue":"ACM Transactions on Software Engineering and Methodology","topic":"Software Testing and Debugging Techniques","field":"Computer Science","cited_by":25,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"National Key Research and Development Program of China; National Natural Science Foundation of China","keywords":"Comet; Computer science; Layer (electronics); Set (abstract data type); Test set; Artificial intelligence; Machine learning; Algorithm; Data mining; Programming language; Chemistry","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":[0.0007956628,0.000184083,0.0002371268,0.0003953178,0.0002916865,0.00008716332,0.0003453709,0.0001400107,0.000001707895],"category_scores_gemma":[0.004808644,0.0001952347,0.00006671017,0.0006426124,0.0000203333,0.0002773966,0.00003505238,0.0002798914,0.000003831813],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001894698,"about_ca_system_score_gemma":0.00003528952,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00000676734,"about_ca_topic_score_gemma":2.703156e-7,"domain_scores_codex":[0.9987992,0.0001234289,0.0002268874,0.0004282833,0.0001010372,0.0003211779],"domain_scores_gemma":[0.9931397,0.006251647,0.00005366887,0.0004113216,0.00005647064,0.00008723693],"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":[0.000004232185,0.00001177925,0.000228106,0.00004546839,0.0000208132,0.000003839242,0.0002697213,0.7225696,0.0009941804,0.0004623552,0.0003956346,0.2749943],"study_design_scores_gemma":[0.0002115371,0.0001430983,0.0001806499,0.0000294681,0.00001481665,0.00004243648,0.000002508626,0.986962,0.002089904,0.009750651,0.0003621203,0.0002108611],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.005185702,0.00008061553,0.9763682,0.0002146351,0.000386009,0.0001625416,0.000005722479,0.01759047,0.000006085505],"genre_scores_gemma":[0.05716598,0.0000573755,0.9421505,0.0001781405,0.00006846637,0.0001474348,0.0000144304,0.00004110225,0.0001765806],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.2747834,"threshold_uncertainty_score":0.796144,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.210945814946218,"score_gpt":0.3372706176748377,"score_spread":0.1263248027286197,"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."}}