{"id":"W6912991590","doi":"10.5281/zenodo.7998401","title":"Towards Foundational Models for Molecular Learning on Large-Scale Multi-Task Datasets","year":2024,"lang":"en","type":"preprint","venue":"RWTH Publications (RWTH Aachen)","topic":"Advanced Graph Neural Networks","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"HEC Montréal; Canadian Institute for Advanced Research; McGill University; Université de Montréal","funders":"","keywords":"Supervised learning; Training set; Graph; Labeled data; Data point; Range (aeronautics); Data type; Code (set theory)","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":[0.005499671,0.002522744,0.002303867,0.002788292,0.001213998,0.003229245,0.005769799,0.002773288,0.002839861],"category_scores_gemma":[0.01807448,0.001343376,0.003722457,0.003166149,0.00152815,0.005868906,0.00458158,0.007485807,0.002669443],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002890581,"about_ca_system_score_gemma":0.003252519,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009961881,"about_ca_topic_score_gemma":0.02212654,"domain_scores_codex":[0.9972203,0.0009995608,0.0001642616,0.0008966374,0.0005057848,0.0002133846],"domain_scores_gemma":[0.9922815,0.003861481,0.0004701549,0.001871381,0.001125378,0.0003900184],"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":[0.000397986,0.0005279648,0.005917673,0.0008550971,0.0004332851,0.0001829033,0.0002242338,0.7374039,0.002993033,0.03508641,0.04372515,0.1722524],"study_design_scores_gemma":[0.00001917761,0.00002932567,0.0002077681,0.00002861478,0.00001322554,0.00001253641,0.00002290829,0.9757529,0.000488926,0.02170211,0.001712806,0.000009775517],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02543481,0.001343089,0.9496165,0.001528428,0.0001510794,0.0002695363,0.007914639,0.0115869,0.002155149],"genre_scores_gemma":[0.2124284,0.001260904,0.734477,0.001270912,0.0002131405,0.001526107,0.04456384,0.001385804,0.002873843],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.009961881,"threshold_uncertainty_score":0.0290854,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04311897580672169,"score_gpt":0.3265490492339211,"score_spread":0.2834300734271994,"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."}}