{"id":"W4402947492","doi":"10.1093/bioinformatics/btae561","title":"RetroCaptioner: beyond attention in end-to-end retrosynthesis transformer via contrastively captioned learnable graph representation","year":2024,"lang":"en","type":"article","venue":"Bioinformatics","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":17,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Young Scientists Fund; National Natural Science Foundation of China; Central South University; Institute for Catastrophic Loss Reduction","keywords":"Computer science; Encoder; Transformer; Artificial intelligence; Graph; Natural language processing; Theoretical computer science; Engineering","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008919376,0.002278712,0.001052394,0.0008961188,0.0004332939,0.001380288,0.003344733,0.001915731,0.01093545],"category_scores_gemma":[0.003476142,0.000615539,0.001432071,0.0007779885,0.0008034575,0.002960967,0.00208203,0.003118777,0.006509633],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001042593,"about_ca_system_score_gemma":0.001513833,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003074282,"about_ca_topic_score_gemma":0.005763424,"domain_scores_codex":[0.999415,0.0001037569,0.00002425865,0.000246264,0.0001539497,0.00005666557],"domain_scores_gemma":[0.9990762,0.0003588381,0.0000691626,0.0002872783,0.0001516171,0.00005677074],"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":[0.0007025927,0.0004387807,0.001593994,0.001016166,0.0001810121,0.0005849199,0.0001991047,0.2143402,0.05210254,0.01926156,0.05823243,0.6513466],"study_design_scores_gemma":[0.00005333527,0.0001451802,0.0001678641,0.0000326746,0.00003701367,0.0001268677,0.00003003347,0.9290181,0.04198037,0.01535401,0.01301917,0.00003541954],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02188494,0.001159615,0.8863968,0.0006359836,0.0003036894,0.0002682044,0.003579679,0.07988552,0.005885703],"genre_scores_gemma":[0.2401832,0.0008848833,0.7208024,0.001118815,0.0001657362,0.0004895445,0.01895936,0.004401577,0.01299451],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01093545,"threshold_uncertainty_score":0.03658277,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01263662405181578,"score_gpt":0.2618396766940532,"score_spread":0.2492030526422374,"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."}}