{"id":"W7115579649","doi":"10.26434/chemrxiv-2025-tpfct","title":"Practical Data-Driven Interrogation of Reactivity in Acid-Catalyzed Carbonyl–Olefin Metathesis with Machine Learning and Large Language Models","year":2025,"lang":"","type":"article","venue":"ChemRxiv","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institutes of Health Research; Research Corporation for Science Advancement","keywords":"Blueprint; Catalysis; Scheduling (production processes); Interpretability; Metathesis; Substrate (aquarium); Bayesian inference; Inference","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":[0.002647069,0.001071771,0.0007730242,0.0006473634,0.00044224,0.001184222,0.001300664,0.00130568,0.001135836],"category_scores_gemma":[0.005664897,0.0004641694,0.001346883,0.000579417,0.0008428419,0.001381165,0.0008364453,0.001923462,0.0007227259],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001047279,"about_ca_system_score_gemma":0.001256385,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004117976,"about_ca_topic_score_gemma":0.006889661,"domain_scores_codex":[0.9991479,0.0003559893,0.00004521603,0.0002542188,0.0001330532,0.00006356701],"domain_scores_gemma":[0.9973668,0.001757323,0.0002103274,0.0003796512,0.0002215314,0.00006435063],"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":[0.0005920214,0.000405257,0.01569079,0.0004880098,0.0002342565,0.0001884134,0.0001268804,0.892644,0.02125444,0.008154494,0.005226083,0.05499525],"study_design_scores_gemma":[0.0000209379,0.00006381007,0.0008691077,0.000007660134,0.00001462286,0.00001718403,0.00002133329,0.9865123,0.007607734,0.004163899,0.000684304,0.00001704328],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8098285,0.001936609,0.1695173,0.001584659,0.00009091522,0.0001119836,0.009441582,0.004246579,0.003241881],"genre_scores_gemma":[0.9346124,0.000352157,0.04956309,0.0002164699,0.00003936424,0.0001381425,0.01387705,0.0001635937,0.001037665],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.004117976,"threshold_uncertainty_score":0.01399922,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02863788593630032,"score_gpt":0.3246500443258162,"score_spread":0.2960121583895159,"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."}}