Knowledge Tree
Relations (1)
related 2.00 — strongly supporting 3 facts
GPT-4o is directly linked to the concept of hallucination through empirical evaluations where it exhibited specific error rates in tasks like Chronological Ordering and Lab Data Understanding [1]. These hallucinations were further analyzed by medical experts to determine their clinical risk severity [2], [3].
Facts (3)
Sources
Medical Hallucination in Foundation Models and Their ... medrxiv.org 2 facts
measurementGPT-4o exhibited the highest hallucination rates in Chronological Ordering (24.6%) and Lab Data Understanding (18.7%) compared to other models, with many of these hallucinations classified by medical experts as posing 'Significant' or 'Considerable' clinical risk.
measurementThe study evaluated hallucination rates and clinical risk severity for five Large Language Models: o1, gemini-2.0-flash-exp, gpt-4o, gemini-1.5-flash, and claude-3.5 sonnet.
Medical Hallucination in Foundation Models and Their Impact on ... medrxiv.org 1 fact
claimMedical experts independently classified a substantial proportion of GPT-4o's hallucinations as posing 'Significant' or 'Considerable' clinical risk.