{"id":"W3101473090","doi":"10.26434/chemrxiv.11869692.v5","title":"RetroXpert: Decompose Retrosynthesis Prediction Like A Chemist","year":2020,"lang":"en","type":"article","venue":"ChemRxiv","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"NeuroDevNet","funders":"Cancer Prevention and Research Institute of Texas","keywords":"Retrosynthetic analysis; Interpretability; Synthon; Computer science; Artificial intelligence; Stepping stone; Margin (machine learning); Process (computing); Machine learning; Engineering; Chemistry; Programming language","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001025486,0.001669098,0.0009135812,0.0007791306,0.0003980854,0.001001752,0.002415361,0.001428214,0.01033574],"category_scores_gemma":[0.002321803,0.0006647923,0.001210702,0.0004142632,0.000573524,0.001703216,0.0009737047,0.001474739,0.003498543],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008081787,"about_ca_system_score_gemma":0.001756548,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002010242,"about_ca_topic_score_gemma":0.004685917,"domain_scores_codex":[0.9996548,0.00004948276,0.00001517977,0.000156111,0.00009813212,0.00002647321],"domain_scores_gemma":[0.9992672,0.000409394,0.00006596618,0.0001486313,0.00006975842,0.0000390405],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001048157,0.0003512731,0.004347761,0.0007723535,0.0002054453,0.0004416858,0.0001538328,0.2808333,0.04004652,0.03193136,0.03651064,0.6033576],"study_design_scores_gemma":[0.0000607914,0.00007264324,0.0001483746,0.00001780878,0.00002659039,0.00006485377,0.00001378907,0.9649989,0.01504453,0.009581432,0.00995399,0.00001637624],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02013981,0.0003823909,0.933218,0.0004910268,0.0001526011,0.0002421495,0.001119215,0.03914411,0.00511073],"genre_scores_gemma":[0.1176208,0.0002683986,0.8719132,0.0004587712,0.00006237007,0.0003083332,0.002063856,0.001972327,0.005332046],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01033574,"threshold_uncertainty_score":0.03457654,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01501172460227548,"score_gpt":0.2342369191532975,"score_spread":0.219225194551022,"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."}}