{"id":"W7130617269","doi":"10.1109/fllm67465.2025.11391072","title":"Energy-Aware Code Generation with LLMs: Benchmarking Small vs. Large Language Models for Sustainable AI Programming","year":2025,"lang":"","type":"article","venue":"","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ericsson (Canada); Algoma University","funders":"","keywords":"Benchmarking; Correctness; Python (programming language); Coding (social sciences); Efficient energy use; Key (lock); Code generation; Compiler","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":[0.002263974,0.001391642,0.0004162724,0.0009022438,0.0003808081,0.001100881,0.002508624,0.001385477,0.004099834],"category_scores_gemma":[0.01461487,0.0005654646,0.001066582,0.001031539,0.0009412313,0.002013853,0.001362437,0.002202907,0.002238211],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001467941,"about_ca_system_score_gemma":0.002657075,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007223395,"about_ca_topic_score_gemma":0.01164064,"domain_scores_codex":[0.9982978,0.0005965682,0.0001379892,0.0003745566,0.0004412861,0.000151763],"domain_scores_gemma":[0.9929178,0.004292338,0.0002754763,0.001177334,0.001098892,0.0002382115],"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":[0.001311476,0.001428373,0.01953493,0.0024179,0.00030771,0.0004686879,0.0009291603,0.5672832,0.01861068,0.01034492,0.0671403,0.3102227],"study_design_scores_gemma":[0.0002733236,0.0003989633,0.002015428,0.0001018848,0.00005883648,0.0001117595,0.0001827131,0.9547948,0.01803576,0.006203355,0.0177785,0.00004482712],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6809375,0.002836218,0.1843934,0.002427418,0.0008479877,0.000997076,0.008385804,0.09324015,0.02593441],"genre_scores_gemma":[0.6916332,0.0009561765,0.2659128,0.001019427,0.00006258892,0.001074028,0.02199387,0.009581728,0.007766109],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.007223395,"threshold_uncertainty_score":0.01436269,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01185613615711992,"score_gpt":0.273740018289398,"score_spread":0.2618838821322781,"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."}}