{"id":"W4408844959","doi":"10.1145/3723876","title":"Automatically Improving LLM-based Verilog Generation using EDA Tool Feedback","year":2025,"lang":"en","type":"article","venue":"ACM Transactions on Design Automation of Electronic Systems","topic":"Software Engineering Research","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"","keywords":"Computer science; Verilog; Computer architecture; Embedded system; Parallel computing; 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001060391,0.0002177909,0.0002898885,0.0006617482,0.000216438,0.00021743,0.0008700136,0.0001566995,0.00001752763],"category_scores_gemma":[0.0004437644,0.0002309016,0.0001056541,0.001213161,0.00003586126,0.0004761198,0.00001294595,0.0002647601,0.00002479285],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007980801,"about_ca_system_score_gemma":0.001074844,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005988227,"about_ca_topic_score_gemma":0.000002685067,"domain_scores_codex":[0.9976088,0.0003044074,0.0006156593,0.0004327024,0.0005510838,0.0004873154],"domain_scores_gemma":[0.9972607,0.001168522,0.0001620886,0.00109042,0.0002636614,0.00005462359],"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":[0.00001841753,0.0001612659,0.00001794727,0.0002204733,0.00009049794,0.000001382816,0.00007444136,0.9007902,0.0661629,0.004773436,0.0001127187,0.02757637],"study_design_scores_gemma":[0.0005316193,0.0002627229,0.000208257,0.0001417266,0.00002418079,0.000007877081,0.000007885174,0.950222,0.04814051,0.0002339122,0.00004343154,0.0001759454],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02141879,0.000158885,0.9761227,0.0001926755,0.0005939996,0.0007999814,0.000002348665,0.0006893464,0.00002127429],"genre_scores_gemma":[0.8697078,0.000003561174,0.1299355,0.00003511231,0.00003010017,0.0001311095,0.000003224009,0.0000183192,0.0001352726],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.848289,"threshold_uncertainty_score":0.9415893,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03001254172624539,"score_gpt":0.2764101130795037,"score_spread":0.2463975713532583,"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."}}