hallucination rate ↔ Large Language Models

Relations (1)

related 0.50 — strongly supporting 5 facts

Large Language Models are the subject of research into their performance metrics, specifically their hallucination rates, as evidenced by studies evaluating these models [1] and frameworks designed to assess their clinical safety {fact:1, fact:2}. Furthermore, research indicates that specific operational factors, such as system instructions, directly influence the hallucination rates observed in these Large Language Models {fact:3, fact:4}.

Facts (5)

Sources

A framework to assess clinical safety and hallucination rates of LLMs ... nature.com Nature 3 facts

claimD.P., E.A., M.D., N.M., S.K., and J.B. contributed to the concept, design, and execution of the study regarding clinical safety and hallucination rates of LLMs.

referenceThe article titled 'A framework to assess clinical safety and hallucination rates of LLMs for medical text summarisation' was published in the journal npj Digital Medicine (volume 8, article 274) in 2025, authored by E. Asgari, N. Montaña-Brown, M. Dubois, and others.

claimThe authors propose a framework for assessing clinical safety and hallucination rates in large language models (LLMs) that includes an error taxonomy for classifying outputs, an experimental structure for iterative comparisons in document generation pipelines, a clinical safety framework to evaluate error harms, and a graphical user interface named CREOLA.

Phare LLM Benchmark: an analysis of hallucination in ... giskard.ai Giskard 1 fact

claimGiskard's data indicates that modifying system instructions significantly impacts the hallucination rates of Large Language Models.

Medical Hallucination in Foundation Models and Their ... medrxiv.org medRxiv 1 fact

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.