{"id":"W4410782593","doi":"10.3390/ai6060109","title":"What We Know About the Role of Large Language Models for Medical Synthetic Dataset Generation","year":2025,"lang":"en","type":"article","venue":"AI","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Fundação para a Ciência e a Tecnologia","keywords":"Computer science; Natural language processing; Data science","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0006112003,0.00005809567,0.00008990823,0.00003888294,0.0001032976,0.0000796263,0.0006795145,0.00005531781,0.00002427331],"category_scores_gemma":[0.0001765175,0.00004085728,0.00002875784,0.0001250843,0.00002143403,0.0002823653,0.0001997996,0.0001371183,0.000006779945],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001323176,"about_ca_system_score_gemma":0.000124166,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001055192,"about_ca_topic_score_gemma":0.00015557,"domain_scores_codex":[0.9991652,0.0001213662,0.0001599225,0.0001897191,0.0002128408,0.0001509796],"domain_scores_gemma":[0.9990951,0.0002135506,0.00004500126,0.0005646,0.00004566168,0.00003612293],"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":[0.000004713355,0.00005558371,0.0001218421,0.0001150517,0.00001750279,0.000002406571,0.003075685,0.001781701,0.0002131543,0.477818,0.02504867,0.4917457],"study_design_scores_gemma":[0.0001014317,0.00001696448,0.00002149744,0.0001036839,0.00000393647,0.00000162092,0.0001095841,0.9075622,0.0006547236,0.004911766,0.08647583,0.0000367641],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003467864,0.02008809,0.9011298,0.07342047,0.0008810275,0.0004527443,0.0002341546,0.00007878055,0.0002470782],"genre_scores_gemma":[0.9869331,0.0006453027,0.004672539,0.006837255,0.0001634523,0.00007865582,0.0003305773,0.000008525766,0.0003306147],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9834652,"threshold_uncertainty_score":0.1666111,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01422989031921169,"score_gpt":0.3408195950236957,"score_spread":0.326589704704484,"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."}}