{"id":"W4413934934","doi":"10.26434/chemrxiv-2025-1lcpq","title":"The elephant in the lab: Synthesizability in generative small-molecule design","year":2025,"lang":"en","type":"article","venue":"ChemRxiv","topic":"Nanotechnology research and applications","field":"Engineering","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canadian Institute for Advanced Research; Vector Institute; University of Toronto","funders":"Natural Resources Canada; Natural Sciences and Engineering Research Council of Canada; University of Toronto; Canada First Research Excellence Fund; Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung; National Science Foundation","keywords":"Generative grammar; Computer science; Artificial intelligence","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005148111,0.0000650525,0.00006854119,0.00004846902,0.00007214712,0.0000245415,0.000348006,0.00006079986,0.000003910694],"category_scores_gemma":[0.0001790969,0.00003986847,0.00001884187,0.0005129695,0.00009099559,0.00001564548,0.00002858773,0.0002445884,0.00001299878],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005123366,"about_ca_system_score_gemma":0.00003363497,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008431557,"about_ca_topic_score_gemma":0.000133153,"domain_scores_codex":[0.9994609,0.00006559051,0.0001127456,0.0001090524,0.00004649127,0.0002052171],"domain_scores_gemma":[0.9992357,0.0004236836,0.000006262904,0.0003090295,0.00001336245,0.00001197259],"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":[0.00004873562,0.0003275446,0.005725371,0.00009872336,0.00009138433,0.00003120708,0.00105208,0.006854499,0.8940623,0.05182847,0.008728249,0.03115139],"study_design_scores_gemma":[0.0001908194,0.000008935564,0.002747203,0.00001752221,0.000003157114,8.177145e-7,0.0002399399,0.01691714,0.9683152,0.008394618,0.003079925,0.0000846945],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.958558,0.001629487,0.03016816,0.005603642,0.00003138217,0.0007524394,8.636409e-7,0.0001232387,0.003132819],"genre_scores_gemma":[0.9988361,0.0001556062,0.0003704761,0.00005999714,0.00000826435,0.0004840202,6.733426e-7,0.000005164763,0.00007967988],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.07425288,"threshold_uncertainty_score":0.1625789,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01952854007483309,"score_gpt":0.2564613348156651,"score_spread":0.236932794740832,"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."}}