{"id":"W7106787043","doi":"10.48448/w4bk-by56","title":"Improving Chemical Understanding of LLMs via SMILES Parsing","year":2025,"lang":"","type":"other","venue":"Open MIND","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Parsing; Comprehension; Construct (python library); Suite; Matching (statistics); Graph; Molecular graph","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":[0.001395536,0.003564698,0.001261772,0.00196792,0.001071679,0.002693846,0.004502734,0.003738619,0.02496005],"category_scores_gemma":[0.00832027,0.001066476,0.003504871,0.001317598,0.001095561,0.00690223,0.002939592,0.005180925,0.01366449],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001589757,"about_ca_system_score_gemma":0.002450958,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006326913,"about_ca_topic_score_gemma":0.01397748,"domain_scores_codex":[0.998735,0.0003395175,0.00006619049,0.0005205724,0.000235439,0.0001032609],"domain_scores_gemma":[0.9972658,0.001543834,0.0001443474,0.0006349555,0.0002956435,0.000115412],"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":[0.000878302,0.001005082,0.005884352,0.002907232,0.0003939084,0.0008860547,0.0004047634,0.1841879,0.02656906,0.03657086,0.3117722,0.4285402],"study_design_scores_gemma":[0.0001966116,0.0002639448,0.0009746767,0.0001555189,0.0001089631,0.0001869676,0.000210501,0.845208,0.02616391,0.06496916,0.06148234,0.00007939463],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1446221,0.006051757,0.4405175,0.00588866,0.001397425,0.0005990131,0.07396256,0.2873805,0.03958041],"genre_scores_gemma":[0.3568614,0.001708345,0.396676,0.003027441,0.0002812463,0.000774866,0.2172877,0.008757723,0.01462522],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02496005,"threshold_uncertainty_score":0.08349967,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07591971258603587,"score_gpt":0.3210920971634381,"score_spread":0.2451723845774022,"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."}}