{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.000495756,0.0003829208,0.0004085464,0.000172926,0.0001880927,0.0001854608,0.001675105,0.0002917983,0.00001235221],"category_scores_gemma":[0.0001141143,0.0003899346,0.0003743689,0.0006367557,0.00004300055,0.0005370976,0.001979721,0.0005331864,0.00002104731],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002689519,"about_ca_system_score_gemma":0.0005074237,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000185335,"about_ca_topic_score_gemma":0.000005875572,"domain_scores_codex":[0.9972091,0.0002592341,0.0002862145,0.00162159,0.0001854753,0.0004383533],"domain_scores_gemma":[0.9981419,0.0002132247,0.0002158554,0.0009133155,0.000279792,0.0002359338],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00009192741,0.0002850884,0.0006179268,0.0005020691,0.0003421789,0.0001377767,0.002945886,0.9055,0.0004601751,0.08591474,0.0007286262,0.002473632],"study_design_scores_gemma":[0.0008228169,0.00008535315,0.000235981,0.00007448326,0.00009986284,0.000003764722,0.00008459933,0.9575139,0.0005491877,0.03899616,0.001119202,0.0004147382],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07098959,0.00003137492,0.9241995,0.0004124051,0.0006176859,0.000965506,0.00009334871,0.0006489404,0.002041712],"genre_scores_gemma":[0.7831627,0.00001797479,0.215934,0.0002608376,0.0001436791,0.00001311706,0.0001013003,0.00003973941,0.0003266121],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7121732,"threshold_uncertainty_score":0.9998553,"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."}}