{"id":"W4400270729","doi":"10.2196/54345","title":"Reference Hallucination Score for Medical Artificial Intelligence Chatbots: Development and Usability Study","year":2024,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":122,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Preprint; Natural language processing; Computer science; Artificial intelligence; Psychology; World Wide Web","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002287183,0.0001740291,0.0002936946,0.000170568,0.0001705447,0.00008983232,0.0001560117,0.0003207776,0.0005024762],"category_scores_gemma":[0.001901462,0.0001344466,0.00004067945,0.0003157897,0.0002067849,0.0002593765,0.00007811557,0.0005337714,0.0001213925],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001764035,"about_ca_system_score_gemma":0.001661348,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004366033,"about_ca_topic_score_gemma":0.0001578846,"domain_scores_codex":[0.9967407,0.00004005818,0.001403123,0.0002300578,0.001258984,0.0003270642],"domain_scores_gemma":[0.9982493,0.0005876599,0.00008656774,0.0002274423,0.000268241,0.0005808239],"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":[0.00007119456,0.0004197488,0.001689908,0.001253798,0.00003109693,0.00001213816,0.02794047,7.428945e-7,0.00000196074,0.004214385,0.0003210326,0.9640436],"study_design_scores_gemma":[0.0004481079,0.00566653,0.01521813,0.007146742,0.0003048735,0.000548917,0.1894936,0.6408246,0.00547763,0.02115306,0.1122024,0.001515502],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9393993,0.0001855594,0.05241342,0.0048122,0.0006997421,0.001852011,0.000003988247,0.0001674605,0.0004663074],"genre_scores_gemma":[0.9932675,0.00008244257,0.00475293,0.0008844984,0.0004138755,0.0004448505,0.00006787958,0.00001450652,0.0000715508],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.962528,"threshold_uncertainty_score":0.5501763,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2320138201162752,"score_gpt":0.4742133288584646,"score_spread":0.2421995087421894,"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."}}