{"id":"W4379115941","doi":"10.23919/date56975.2023.10137086","title":"Benchmarking Large Language Models for Automated Verilog RTL Code Generation","year":2023,"lang":"en","type":"article","venue":"","topic":"Software Engineering Research","field":"Computer Science","cited_by":142,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"Army Research Office; National Science Foundation","keywords":"Computer science; Verilog; Programming language; Compiler; Correctness; Construct (python library); Scripting language; Embedded system; Field-programmable gate array","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004708408,0.002116857,0.0005593232,0.001669911,0.0005733717,0.001639907,0.003479681,0.001847438,0.007289788],"category_scores_gemma":[0.02597914,0.001135916,0.001663032,0.001273602,0.001138453,0.003456198,0.001982841,0.003038118,0.004967199],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00213026,"about_ca_system_score_gemma":0.003600415,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007382188,"about_ca_topic_score_gemma":0.01473274,"domain_scores_codex":[0.9957303,0.001530465,0.0004106837,0.0009045813,0.001088966,0.000334906],"domain_scores_gemma":[0.9833502,0.009503154,0.0007678891,0.003517256,0.002530083,0.0003314691],"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":[0.001384405,0.001684454,0.0206165,0.003535984,0.0004647586,0.0005413949,0.0007771144,0.4030132,0.02941464,0.01377324,0.1505856,0.3742088],"study_design_scores_gemma":[0.0002921838,0.0004404797,0.002546144,0.0002266006,0.00008273758,0.0002115435,0.0001853943,0.9259245,0.03248213,0.00840903,0.0291274,0.00007190147],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.340223,0.004060035,0.3768925,0.002585333,0.00122279,0.001479225,0.03000676,0.2253238,0.01820654],"genre_scores_gemma":[0.4897899,0.001166996,0.399184,0.001266178,0.000105621,0.001270021,0.08678579,0.0144819,0.005949599],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007382188,"threshold_uncertainty_score":0.02490073,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04206808642623612,"score_gpt":0.3182129701439342,"score_spread":0.2761448837176981,"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."}}