{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","sts","scholarly_communication","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.003746108,0.0007003811,0.0007730639,0.0004096279,0.00178698,0.003080237,0.001163175,0.0003151071,0.0009493565],"category_scores_gemma":[0.0002107261,0.0005994256,0.000145505,0.001008657,0.0002753235,0.001526432,0.0007088719,0.0002599677,0.00001203369],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003915114,"about_ca_system_score_gemma":0.001025379,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002070456,"about_ca_topic_score_gemma":0.002080946,"domain_scores_codex":[0.9939255,0.0004651491,0.0009851174,0.001772418,0.0006611713,0.002190711],"domain_scores_gemma":[0.9970494,0.0002359776,0.0004930298,0.001001544,0.0009870506,0.0002330279],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001186327,0.0009796807,0.0007035611,0.004484778,0.000140841,0.0003016607,0.00646584,0.3242713,0.1785685,0.4586793,0.005663984,0.01855428],"study_design_scores_gemma":[0.001384131,0.0005754783,0.000009474922,0.0003222241,0.0001236672,0.00001805904,0.002942218,0.9122453,0.06693592,0.00128362,0.01341929,0.0007406699],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05195916,0.0003408648,0.9394798,0.001116123,0.001011567,0.001761774,0.00006870108,0.0003399748,0.003922076],"genre_scores_gemma":[0.8472279,0.00002755134,0.1213292,0.001947113,0.0005675568,0.0006545794,0.0001537784,0.00007531611,0.02801705],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8181506,"threshold_uncertainty_score":0.9999639,"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."}}