{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.02017564,0.0007976218,0.001380628,0.003893672,0.0005761299,0.001475772,0.0007763566,0.0007372507,0.001427689],"category_scores_gemma":[0.05820855,0.0004005887,0.001846833,0.001396376,0.0008301558,0.001642937,0.002094376,0.0007502132,0.0003732869],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001046861,"about_ca_system_score_gemma":0.0007850287,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001085591,"about_ca_topic_score_gemma":0.001741023,"domain_scores_codex":[0.9861464,0.00721212,0.002206359,0.0009930948,0.002995448,0.0004466016],"domain_scores_gemma":[0.935267,0.04452989,0.003761663,0.002278612,0.01204841,0.002114418],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.006014195,0.007148585,0.5135186,0.005660446,0.001091102,0.001209699,0.04797526,0.002750368,0.02115895,0.0008399196,0.00556716,0.3870658],"study_design_scores_gemma":[0.0009925142,0.03059556,0.8960693,0.0008399382,0.001180673,0.002276234,0.01859554,0.024355,0.01381306,0.0008668157,0.009987402,0.0004280222],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9933251,0.0002232746,0.003515361,0.00007741579,0.00003220016,0.001176377,0.0002247575,0.0002267477,0.00119878],"genre_scores_gemma":[0.9825459,0.0002025548,0.01395602,0.00005765334,0.00002996818,0.001748103,0.0005722591,0.00006140018,0.0008262671],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9798244,"threshold_uncertainty_score":0.1067002,"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."}}