Knowledge Tree
Relations (1)
related 1.00 — strongly supporting 1 fact
The concept of hallucination is related to LLaMA because prompt design has been shown to influence hallucination rates specifically within the LLaMA 2 model, as stated in [1].
Facts (1)
Sources
Survey and analysis of hallucinations in large language models frontiersin.org 1 fact
claimPrompt design strongly influences hallucination rates in prompt-sensitive models such as LLaMA 2 and OpenChat.