Ethan Mollick Prefers Personalized LLM Explanations Over ELI5
He recommends asking LLMs to draw on users existing knowledge rather than using generic ELI5 prompts.
Associate Professor Ethan Mollick posted that he dislikes ELI5 as the default prompt for simplifying topics. He argues that research shows drawing connections to what a person already knows improves memory and context building. Mollick states that users are not children and do not need things dumbed down. Instead he favors prompts that ask an LLM to explain a topic using the users prior knowledge. Engineer Thariq Shihipar replied that the ELI5 label is misleading because the prompt actually seeks an explanation without assuming domain expertise. The exchange appears among visible posts discussing LLM prompting techniques.
I don’t like ELI5 as default for “make this easy to understand,” a great thing about LLMs is the ability to work as universal translators across topics Asking for generic dumbed-down explanations is worse than asking for personalized ones: “explain to me drawing on what I know”
a skill people at Anthropic have been using a lot recently: ELI5 /eli5 <what you want explained> "explain like I'm someone who knows nothing about this topic, using a HTML artifact with big pictures and few words"
Reactions from ranked influencers
3 postsResearch shows that drawing connections to what you already know helps you remember and build context. And you aren’t five! You don’t need things dumbed down, you need things explained differently. The fact that LLMs can actually do this is one of their most exciting features
@emollick eli5 is maybe too clickbaity of a name, the prompt doesn’t actually ask you to explain it like I’m 5, just like you don’t have domain knowledge I think the “big pictures, few words” part of that prompt is actually the most important part
ELI5 —> ELIM: explain like i’m me
I don’t like ELI5 as default for “make this easy to understand,” a great thing about LLMs is the ability to work as universal translators across topics Asking for generic dumbed-down explanations is worse than asking for personalized ones: “explain to me drawing on what I know”…
