{"id":"W4410253554","doi":"10.1016/j.commatsci.2025.113919","title":"Small-cell-based fast active learning of machine learning interatomic potentials","year":2025,"lang":"en","type":"article","venue":"Computational Materials Science","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Nuclear Laboratories; Queen's University","funders":"Alliance de recherche numérique du Canada; University Network of Excellence in Nuclear Engineering; Atomic Energy of Canada Limited; Natural Sciences and Engineering Research Council of Canada; Mitacs","keywords":"Interatomic potential; Active learning (machine learning); Materials science; Computer science; Molecular dynamics; Chemistry; Artificial intelligence; Computational chemistry","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.002777954,0.0003582302,0.0005790783,0.0007157681,0.00100635,0.0006411581,0.001507265,0.00009649921,0.001340617],"category_scores_gemma":[0.001112412,0.0003386611,0.00007388237,0.001249278,0.001144433,0.0005263066,0.0006451936,0.0002687037,0.000184629],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001793139,"about_ca_system_score_gemma":0.0008192736,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002918869,"about_ca_topic_score_gemma":0.000006110677,"domain_scores_codex":[0.9957868,0.0006613052,0.0009479432,0.0009974363,0.0009341268,0.0006723581],"domain_scores_gemma":[0.9974818,0.0005494552,0.0007859711,0.0003618988,0.0006789946,0.0001419403],"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":[0.00007972657,0.00007055166,0.0003056522,0.00009240251,0.000002578502,0.000002532456,0.0002494505,0.3954165,0.6025357,0.0009988208,0.00000877417,0.0002373654],"study_design_scores_gemma":[0.0007255698,0.0001698615,0.003169568,0.0002083718,0.00001431475,0.000007174409,0.0001790608,0.1496753,0.8442471,0.001228526,0.00009166171,0.0002834531],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8994103,0.00003713726,0.09738708,0.0001701978,0.001321427,0.0003281194,0.00003929764,0.0002171165,0.001089293],"genre_scores_gemma":[0.9795337,0.000002724609,0.01979828,0.000179293,0.00005772732,0.00003420117,0.00003999089,0.00002443381,0.0003296059],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2457412,"threshold_uncertainty_score":0.9999065,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01061989196915822,"score_gpt":0.2650434841600711,"score_spread":0.2544235921909128,"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."}}