{"id":"W3093934881","doi":"10.48550/arxiv.2010.09885","title":"ChemBERTa: Large-Scale Self-Supervised Pretraining for Molecular Property Prediction","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Computational Drug Discovery Methods","field":"Computer Science","cited_by":397,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Transformer; PubChem; Artificial intelligence; Machine learning; Visualization; Property (philosophy); Transfer of learning; Data mining; Engineering","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":[0.0009866675,0.002059793,0.0008697116,0.001117912,0.0005057961,0.0007743561,0.003274268,0.001950884,0.01007058],"category_scores_gemma":[0.003051928,0.000737438,0.001356537,0.00120661,0.0005893644,0.00157402,0.001208714,0.002651209,0.005176984],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001159761,"about_ca_system_score_gemma":0.001539255,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007050487,"about_ca_topic_score_gemma":0.01761499,"domain_scores_codex":[0.9995458,0.000107414,0.00002158879,0.0001661268,0.0001138764,0.00004522968],"domain_scores_gemma":[0.9989461,0.0005037724,0.00007453088,0.0002638834,0.0001503908,0.0000612636],"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":[0.001057436,0.0010705,0.006002001,0.001422314,0.000548252,0.0003047186,0.0001060849,0.3471529,0.01514769,0.006569473,0.226552,0.3940667],"study_design_scores_gemma":[0.00008925419,0.0001628887,0.0005966335,0.00002831051,0.00003521614,0.00004616535,0.00001801016,0.9752762,0.009825551,0.004386223,0.009519281,0.00001635176],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1369401,0.004535308,0.5576211,0.002184781,0.0008918677,0.0008806116,0.03775743,0.2421159,0.01707291],"genre_scores_gemma":[0.4397095,0.001347388,0.4282233,0.001839728,0.0001784667,0.001530895,0.1065121,0.004970198,0.01568833],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01007058,"threshold_uncertainty_score":0.0336895,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07861375255540573,"score_gpt":0.2193217133781824,"score_spread":0.1407079608227767,"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."}}