{"id":"W4404875179","doi":"10.1016/j.engappai.2024.109490","title":"Low-cost language models: Survey and performance evaluation on Python code generation","year":2024,"lang":"en","type":"article","venue":"Engineering Applications of Artificial Intelligence","topic":"Topic Modeling","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":false,"ca_institutions":"Novelis (Canada)","funders":"","keywords":"Computer science; Python (programming language); Programming language; Code generation; Operating system","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":[0.003346954,0.001897625,0.001023602,0.002412604,0.0008036853,0.002386196,0.007176474,0.001064665,0.009950224],"category_scores_gemma":[0.02136119,0.001372321,0.002024139,0.003760544,0.000954703,0.005083388,0.002585392,0.002659697,0.007281977],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001833193,"about_ca_system_score_gemma":0.004926011,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01481657,"about_ca_topic_score_gemma":0.01435265,"domain_scores_codex":[0.9957371,0.001064403,0.0004318456,0.0005308118,0.001874764,0.0003610472],"domain_scores_gemma":[0.9887721,0.005024893,0.000559446,0.00322157,0.002023733,0.0003982521],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001535199,0.0009962524,0.009366813,0.004239808,0.0004839702,0.0003087229,0.0006250876,0.06688216,0.01418651,0.01870683,0.1569025,0.7257661],"study_design_scores_gemma":[0.0008292026,0.0008169082,0.004090322,0.0007412852,0.0005289641,0.000744704,0.0003686918,0.7636869,0.05928018,0.02307039,0.145593,0.0002495069],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08140396,0.008260959,0.4729638,0.001743464,0.0008683954,0.001213274,0.00892823,0.4055501,0.01906774],"genre_scores_gemma":[0.317499,0.00719536,0.5761889,0.001476893,0.0001873629,0.001062508,0.03951918,0.04724335,0.009627528],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01481657,"threshold_uncertainty_score":0.03328681,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08149017670230732,"score_gpt":0.3154793380580967,"score_spread":0.2339891613557894,"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."}}