{"id":"W4399165032","doi":"10.1021/acs.jcim.3c02070","title":"Application of Transformers in Cheminformatics","year":2024,"lang":"en","type":"article","venue":"Journal of Chemical Information and Modeling","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":56,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Division of Materials Research; Johns Hopkins University; Institute for Catastrophic Loss Reduction; National Science Foundation","keywords":"Computer science; Chemical space; Automatic summarization; Cheminformatics; Artificial intelligence; Machine learning; Transformer; Data science; Drug discovery; Chemistry; 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":[0.003136322,0.0008389853,0.0007294495,0.002572287,0.0007434138,0.002929518,0.002094078,0.001148707,0.006392696],"category_scores_gemma":[0.01082028,0.0005469819,0.001296981,0.003620239,0.002143174,0.00628972,0.003501017,0.002416212,0.003228898],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001521248,"about_ca_system_score_gemma":0.002090483,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0012905,"about_ca_topic_score_gemma":0.001345277,"domain_scores_codex":[0.9980991,0.0006394156,0.0001590413,0.0003549453,0.0006607184,0.00008676002],"domain_scores_gemma":[0.9969637,0.001434504,0.0001788065,0.0008102161,0.0004735517,0.0001391744],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00008932836,0.00005774093,0.001213801,0.0008805025,0.00007777246,0.0002272317,0.0002115632,0.03860706,0.005414354,0.5322583,0.008758646,0.4122037],"study_design_scores_gemma":[0.00002841606,0.00008070748,0.0002332048,0.000205166,0.00005546938,0.0004838252,0.0001141687,0.1676343,0.0108718,0.7170856,0.1031651,0.00004218029],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006526984,0.006655019,0.9639021,0.002726069,0.0004115416,0.0001232027,0.0004417113,0.002658913,0.01655445],"genre_scores_gemma":[0.2264674,0.0198124,0.7423115,0.001482322,0.0005990603,0.0002840335,0.001180252,0.0008409997,0.007022077],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006392696,"threshold_uncertainty_score":0.02138567,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01087750625260464,"score_gpt":0.2736177912065863,"score_spread":0.2627402849539817,"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."}}