{"id":"W4412943124","doi":"10.26434/chemrxiv-2025-wftzz","title":"Underexplored Catalysts as General Structures: Application of Machine Learning Techniques for Reaction-Specific Datasets","year":2025,"lang":"en","type":"preprint","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; University of British Columbia; Canada Foundation for Innovation; Compute Canada","keywords":"Computer science; Catalysis; Artificial intelligence; Machine learning; Chemistry; Organic 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.007830712,0.001643593,0.001026928,0.00345061,0.0007879643,0.002370712,0.002041892,0.001949934,0.001207133],"category_scores_gemma":[0.01707317,0.0004006445,0.001786632,0.002363955,0.001172543,0.002105109,0.001649976,0.002317156,0.0008780629],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009738991,"about_ca_system_score_gemma":0.001315636,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00194941,"about_ca_topic_score_gemma":0.003405702,"domain_scores_codex":[0.9970244,0.001265411,0.0002512573,0.0009105291,0.0004281836,0.0001201925],"domain_scores_gemma":[0.9920258,0.005141052,0.000657963,0.001638016,0.0004142761,0.0001228706],"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":[0.001362051,0.001207778,0.1169166,0.003225865,0.001683053,0.0006862249,0.0005193612,0.5864788,0.02678343,0.01933257,0.02336183,0.2184424],"study_design_scores_gemma":[0.000068456,0.0002469217,0.006545968,0.0001113978,0.0001351429,0.0001943172,0.0001766823,0.9335918,0.0147042,0.02835651,0.0158089,0.00005969245],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.5225272,0.01024314,0.4007997,0.003767029,0.0004642389,0.0005932286,0.04274751,0.01161648,0.007241471],"genre_scores_gemma":[0.7417668,0.001728099,0.2067725,0.00079562,0.0001835291,0.000631418,0.04618454,0.0005985677,0.001338929],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.007830712,"threshold_uncertainty_score":0.04141325,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02323471202680713,"score_gpt":0.3114077971173594,"score_spread":0.2881730850905523,"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."}}