{"id":"W6911194615","doi":"10.5281/zenodo.10104139","title":"Fine-Tuned Large Language Models for Symptom Recognition from Spanish Clinical Text","year":2023,"lang":"en","type":"article","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University","funders":"","keywords":"Categorization; Task (project management); Set (abstract data type); Generative grammar; Unified Medical Language System; Natural language; Health care; Data curation; Health professionals","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":[0.004058,0.00198232,0.001188413,0.001800276,0.0006256042,0.001548309,0.00178331,0.001692904,0.003006266],"category_scores_gemma":[0.01171319,0.000704738,0.001892301,0.001111076,0.0006638545,0.001393187,0.001425929,0.002925381,0.002137639],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001882564,"about_ca_system_score_gemma":0.001760251,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02803964,"about_ca_topic_score_gemma":0.02812156,"domain_scores_codex":[0.9982045,0.0008995533,0.0001231809,0.0005369434,0.0001057627,0.0001300818],"domain_scores_gemma":[0.9878755,0.01029873,0.0002834434,0.000533913,0.0007953002,0.0002131559],"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":[0.002924872,0.001700071,0.01626616,0.0009714182,0.001064198,0.001279812,0.0008880998,0.5561416,0.01808622,0.002719513,0.0376555,0.3603027],"study_design_scores_gemma":[0.00011413,0.00008956259,0.001355035,0.00002016893,0.00008454156,0.00005198862,0.00006438541,0.993735,0.001714989,0.001591115,0.001154998,0.00002416771],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7591293,0.006031861,0.1847664,0.003679079,0.0009761752,0.0006046614,0.01634279,0.02369804,0.004771672],"genre_scores_gemma":[0.9150201,0.0005209922,0.05584059,0.0005987291,0.0002829043,0.0003997182,0.02409691,0.000681666,0.002558346],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02803964,"threshold_uncertainty_score":0.05575287,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08905178192975607,"score_gpt":0.3355385511252263,"score_spread":0.2464867691954702,"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."}}