{"id":"W4226197958","doi":"10.48550/arxiv.2204.00056","title":"SELFIES and the future of molecular string representations","year":2022,"lang":"en","type":"preprint","venue":"White Rose Research Online (University of Leeds, The University of Sheffield, University of York)","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":23,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; University of Toronto; Austrian Science Fund; Coordenação de Aperfeiçoamento de Pessoal de Nível Superior; Fundação de Amparo à Pesquisa do Estado de São Paulo; Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung; National Science Foundation","keywords":"Cheminformatics; Computer science; Artificial intelligence; Interpretability; Popularity; String (physics); Context (archaeology); Representation (politics); Interpretation (philosophy); Interface (matter); Data science; Cognitive science; Chemistry; Programming language; Mathematics; Computational chemistry","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","sts","insufficient_payload"],"consensus_categories":["sts"],"category_scores_codex":[0.003508741,0.0002797007,0.000867784,0.0006175934,0.001465811,0.00003324894,0.004104057,0.0003211169,0.002381339],"category_scores_gemma":[0.0001642816,0.0002915779,0.000385569,0.000897738,0.005827637,0.000275657,0.007560094,0.001488562,0.000005348962],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001432522,"about_ca_system_score_gemma":0.0005630571,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01873743,"about_ca_topic_score_gemma":0.002506955,"domain_scores_codex":[0.9949688,0.002077304,0.0002940928,0.0007217594,0.001478427,0.0004596216],"domain_scores_gemma":[0.9957691,0.0009743857,0.0009518741,0.001343653,0.0008066836,0.0001543409],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"qualitative","study_design_scores_codex":[0.01686619,0.002882396,0.01893964,0.008098256,0.002212107,0.0009295923,0.3482502,0.2606032,0.2475923,0.06881309,0.01603936,0.008773734],"study_design_scores_gemma":[0.009535724,0.001038513,0.03031892,0.001302291,0.001291285,0.0000386508,0.9003514,0.02014168,0.004114641,0.004613965,0.02591964,0.001333247],"study_design_candidate":"qualitative","study_design_consensus":"qualitative","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9839238,0.0007704072,0.003230475,0.007907655,0.0002250244,0.0008075138,0.0007249186,0.00005008275,0.002360113],"genre_scores_gemma":[0.9807947,0.00303968,0.01286686,0.00001339137,0.00004875596,5.256358e-8,0.00007795458,0.00001852876,0.003140101],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5521013,"threshold_uncertainty_score":0.9999536,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02421626748547631,"score_gpt":0.2747418245114769,"score_spread":0.2505255570260005,"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."}}