{"id":"W4416131921","doi":"10.48550/arxiv.2506.04439","title":"RETRO SYNFLOW: Discrete Flow Matching for Accurate and Diverse Single-Step Retrosynthesis","year":2025,"lang":"en","type":"preprint","venue":"ArXiv.org","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Engineering and Physical Sciences Research Council; Alliance de recherche numérique du Canada; Canadian Institute for Advanced Research; Vector Institute; University of British Columbia; Government of Canada","keywords":"Oracle; Inference; Set (abstract data type); Matching (statistics); Resampling; Identification (biology); Retrosynthetic analysis; Product (mathematics); Metric (unit)","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.002124968,0.000680472,0.0009571073,0.0002177586,0.0006578015,0.0007654343,0.001469815,0.0004554143,0.0005279271],"category_scores_gemma":[0.002241626,0.0006227217,0.0002256839,0.0001935683,0.0003683986,0.0004134682,0.003021388,0.0006080006,0.0001475088],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001837103,"about_ca_system_score_gemma":0.00017618,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000420161,"about_ca_topic_score_gemma":0.00006893105,"domain_scores_codex":[0.9955727,0.0004345405,0.0008108413,0.001774482,0.0005326515,0.0008747821],"domain_scores_gemma":[0.9968179,0.0007694953,0.0006877019,0.001319638,0.0001906772,0.0002145822],"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":[0.000433811,0.0001122564,0.04918057,0.00286505,0.0001006601,0.00004859166,0.001665375,0.01476506,0.926486,0.0003965074,0.001770609,0.002175544],"study_design_scores_gemma":[0.004183213,0.001152333,0.163384,0.01111498,0.00219868,0.0000864063,0.001682296,0.207361,0.5640186,0.01511745,0.0205989,0.009102118],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9629425,0.0001861609,0.0287063,0.0009145394,0.004305956,0.0009536702,0.00081077,0.0004217193,0.0007584111],"genre_scores_gemma":[0.9256705,0.00006972563,0.07110826,0.0003798072,0.000530053,0.0001911591,0.00007062432,0.00006963382,0.001910242],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3624673,"threshold_uncertainty_score":0.9996224,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04353494323558441,"score_gpt":0.2978402552627619,"score_spread":0.2543053120271774,"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."}}