{"id":"W4401183144","doi":"10.26434/chemrxiv-2024-mm31v","title":"Assessment of Fine-Tuned Large Language Models for Real-World Chemistry and Material Science Applications","year":2024,"lang":"en","type":"preprint","venue":"ChemRxiv","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; University of Waterloo","funders":"","keywords":"Benchmark (surveying); Set (abstract data type); Computer science; Range (aeronautics); Computation; Work (physics); Simple (philosophy); Fine-tuning; Machine learning; Artificial intelligence; Algorithm; Materials science; Epistemology; Engineering; Geography","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"],"consensus_categories":[],"category_scores_codex":[0.002366196,0.0003514722,0.0005557257,0.0001306458,0.0002281973,0.0005190636,0.001202299,0.0001705033,0.0005413436],"category_scores_gemma":[0.0001015014,0.0003258995,0.00009543497,0.000344468,0.0006862318,0.0001386412,0.002592517,0.0003002721,0.00001393227],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001501349,"about_ca_system_score_gemma":0.0006810194,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001635621,"about_ca_topic_score_gemma":0.00001901941,"domain_scores_codex":[0.9970015,0.00003882727,0.0006046031,0.001255736,0.0005569858,0.0005423017],"domain_scores_gemma":[0.9980615,0.0001061313,0.0004368305,0.001025857,0.000213333,0.0001563736],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00001108786,0.00005047956,0.00005595024,0.002407457,0.000006234246,0.000001643967,0.0001905618,0.001214935,0.9889605,0.006887401,0.0001127291,0.0001009981],"study_design_scores_gemma":[0.0002248498,0.00001699851,0.0004269903,0.0002530611,0.00006336864,0.000003602545,0.0000540677,0.05175928,0.9276026,0.0189488,0.0002628658,0.0003834546],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9775054,0.0001122113,0.01079189,0.0001884566,0.0009726763,0.001127873,0.0005699759,0.0002678888,0.008463638],"genre_scores_gemma":[0.9420763,0.00002178161,0.0554916,0.0000224286,0.0003875764,0.0008141311,0.0001091023,0.00004842376,0.001028677],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.06135786,"threshold_uncertainty_score":0.9999193,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01532537674670113,"score_gpt":0.3398288895007431,"score_spread":0.324503512754042,"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."}}