News for hackers. Stuff that matters.
Re:GPT-4.1 in the API by laggyluke on Wednesday April 16, 2025 @02:15AM (#43702065)
If you're asking an LLM about a particular text, even if it's a well-known text, you might get significantly better results if you provide said text as part of your prompt (context) instead of asking a model to "recall it from memory".

So something like this: "Here's a PDF file containing Being and Time. Please explain the significance of anxiety (Angst) in the uncovering of Being."

There are no comments on this comment yet.