{"id":"W4401312022","doi":"10.18260/1-2--48541","title":"Work-in-Progress: Fine-Tuning Large Language Models for Automated Feedback in Complex Engineering Problem-Solving","year":2024,"lang":"en","type":"article","venue":"","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"3v Geomatics (Canada); Queen's University; University of Alberta","funders":"Queen's University","keywords":"Computer science; Process (computing); Domain (mathematical analysis); Work in process; Fine-tuning; Artificial intelligence; Quality (philosophy); Diversity (politics); Engineering; Programming language","routes":{"ca_aff":true,"ca_fund":true,"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":[0.007196107,0.00221152,0.001111943,0.0009289777,0.0006382595,0.002424683,0.002567478,0.001828679,0.003449504],"category_scores_gemma":[0.02982073,0.0006221294,0.001171543,0.0006708106,0.0009731501,0.004073212,0.002791551,0.00398788,0.002461722],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001176977,"about_ca_system_score_gemma":0.001984989,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004199631,"about_ca_topic_score_gemma":0.006081542,"domain_scores_codex":[0.9959723,0.002419317,0.0001931131,0.0007962775,0.0004272642,0.0001916331],"domain_scores_gemma":[0.9854893,0.009884381,0.0005181382,0.002379721,0.001246436,0.0004820378],"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":[0.001304789,0.001851534,0.008226516,0.0006779611,0.0003184497,0.0001629327,0.0009977926,0.3309431,0.01789982,0.004161495,0.01440024,0.6190554],"study_design_scores_gemma":[0.0001237016,0.0003636791,0.0008653709,0.00004977097,0.00004548862,0.0000483891,0.0001748771,0.9748923,0.01030665,0.007360711,0.005722892,0.00004622506],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.2856351,0.003270095,0.662882,0.00299331,0.0008369111,0.0005783013,0.001590199,0.03409442,0.008119541],"genre_scores_gemma":[0.6944478,0.0004181758,0.2959129,0.0009457602,0.0001610078,0.0004806341,0.003393375,0.00108108,0.003159271],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.007196107,"threshold_uncertainty_score":0.03805715,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01765739101056635,"score_gpt":0.2900544829488644,"score_spread":0.2723970919382981,"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."}}