{"id":"W4403320045","doi":"10.1101/2024.10.07.617002","title":"Improving the reliability of molecular string representations for generative chemistry","year":2024,"lang":"en","type":"preprint","venue":"bioRxiv (Cold Spring Harbor Laboratory)","topic":"Scientific Computing and Data Management","field":"Decision Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"Grand Équipement National De Calcul Intensif; Agence Nationale de la Recherche","keywords":"Generative grammar; String (physics); Reliability (semiconductor); Chemistry; Computer science; Computational biology; Computational chemistry; Theoretical physics; Artificial intelligence; Biology; Physics; Thermodynamics","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":[0.004274931,0.0007177182,0.0008520863,0.0010812,0.0004229131,0.001359071,0.001484717,0.001297933,0.002623274],"category_scores_gemma":[0.02491733,0.0005969959,0.0009906745,0.0005991263,0.001209738,0.002342383,0.002246436,0.002168668,0.0008976185],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008707577,"about_ca_system_score_gemma":0.0007622982,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0014795,"about_ca_topic_score_gemma":0.001301511,"domain_scores_codex":[0.9979948,0.001059243,0.0001094177,0.0002704951,0.0004516079,0.0001144525],"domain_scores_gemma":[0.9829626,0.01233037,0.0009698962,0.002522158,0.0009886102,0.0002263021],"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":[0.0002185299,0.000087835,0.003665866,0.0001121998,0.00005435698,0.00009646622,0.0001350829,0.8840007,0.006925695,0.02139054,0.00135699,0.08195563],"study_design_scores_gemma":[0.000004514539,0.0000197055,0.00009694838,0.000006736878,0.000003472981,0.000009201847,0.000006198311,0.9935268,0.001930771,0.004178922,0.0002122142,0.000004535542],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2227978,0.000375488,0.7680067,0.0006459315,0.0001022098,0.00008948268,0.0003446812,0.004347045,0.003290659],"genre_scores_gemma":[0.9109188,0.0001456202,0.08664674,0.000174621,0.00004458403,0.0001141434,0.0006034414,0.0003966189,0.0009553044],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004274931,"threshold_uncertainty_score":0.02260828,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04636439484916314,"score_gpt":0.315680954438788,"score_spread":0.2693165595896249,"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."}}