{"id":"W4413357328","doi":"10.26434/chemrxiv-2025-c6rxp-v3","title":"Distilling System Complexity to Enable Unbiased and Predictive Computational Reaction Investigations","year":2025,"lang":"en","type":"article","venue":"ChemRxiv","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Sherbrooke","funders":"Fonds de recherche du Québec – Nature et technologies; Université de Sherbrooke; Natural Sciences and Engineering Research Council of Canada; Alliance de recherche numérique du Canada; Hydro-Québec; Centre in Green Chemistry and Catalysis","keywords":"Computer science; Computational complexity theory; Algorithm","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":[0.001175287,0.0009381066,0.001077787,0.0007158045,0.000610217,0.001380449,0.001320683,0.001188632,0.001994162],"category_scores_gemma":[0.004428572,0.000635561,0.001122425,0.0004125418,0.001176028,0.001945446,0.001249734,0.001767828,0.0003915407],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001069359,"about_ca_system_score_gemma":0.001580092,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002826966,"about_ca_topic_score_gemma":0.002998708,"domain_scores_codex":[0.9995647,0.0001753804,0.000020477,0.00008682992,0.0001082082,0.00004438299],"domain_scores_gemma":[0.9979631,0.001486116,0.0001260038,0.0002958457,0.00007873138,0.00005026694],"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":[0.00002113704,0.00002083736,0.0007090331,0.00008170382,0.00002529344,0.00005287808,0.00003883777,0.970981,0.002038607,0.0220773,0.0001789327,0.003774448],"study_design_scores_gemma":[0.000003485178,0.000006813971,0.00006034112,0.000004055667,0.000003506889,0.000006067283,0.000005899086,0.9872441,0.0004898132,0.01184169,0.0003309583,0.000003311032],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1674166,0.000590138,0.8199001,0.001031428,0.00006828566,0.00014441,0.0007006542,0.0009652006,0.009183189],"genre_scores_gemma":[0.8407311,0.0006948954,0.1558325,0.0001661055,0.00005071656,0.0004451417,0.0005917854,0.0002401183,0.001247778],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002826966,"threshold_uncertainty_score":0.007758737,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02452473904853978,"score_gpt":0.2743451841579776,"score_spread":0.2498204451094378,"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."}}