{"id":"W4416035446","doi":"10.18653/v1/2025.emnlp-main.1398","title":"Trustworthy Medical Question Answering: An Evaluation-Centric Survey","year":2025,"lang":"","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Science Foundation of Shandong Province; Natural Sciences and Engineering Research Council of Canada; National Natural Science Foundation of China; Canadian Institute for Advanced Research","keywords":"Trustworthiness; Natural (archaeology); Empirical research; Questions and answers; Natural language","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","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.01285877,0.0002989688,0.0003541641,0.0003994989,0.0002967763,0.0004881706,0.001691955,0.0003595677,0.003119035],"category_scores_gemma":[0.002416622,0.0003022573,0.00008950887,0.001820315,0.00008828488,0.0008998386,0.0005378455,0.0004381785,0.0001356834],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002876288,"about_ca_system_score_gemma":0.002853839,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001956621,"about_ca_topic_score_gemma":0.001832809,"domain_scores_codex":[0.9928851,0.002218921,0.0009070236,0.001221897,0.002155577,0.000611494],"domain_scores_gemma":[0.9967566,0.0003954725,0.0001400448,0.001410655,0.0008553222,0.0004419609],"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":[0.00001669043,0.0002920512,0.05396824,0.00003307455,0.00004193785,0.000009833913,0.0003186536,0.0055745,0.000004249662,0.05659165,0.0008301071,0.882319],"study_design_scores_gemma":[0.0007479166,0.00006289264,0.09199733,0.0001128093,0.00003394356,0.000005372012,0.00002096118,0.9034876,0.00004361364,0.002460923,0.0007700205,0.0002566439],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07774288,0.0009795864,0.9052587,0.002396732,0.004387957,0.000519561,0.000001959183,0.0001812727,0.00853132],"genre_scores_gemma":[0.9870642,0.0001479665,0.009372113,0.001130566,0.000234279,0.0000232832,0.00002300723,0.00001080776,0.001993799],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9093213,"threshold_uncertainty_score":0.999943,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05583913506053498,"score_gpt":0.3540603383243048,"score_spread":0.2982212032637698,"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."}}