3 ms·
This seems a result of relying on the LLM to accurately extract information that needs to be exact. I touch on this from my own experience: https://youtu.be/cs
by Edmond 2y ago
This seems a result of relying on the LLM to accurately extract information that needs to be exact.
I touch on this from my own experience:
https://youtu.be/cs5cbxDClbM?si=IQIFAD38cVzLCs55&t=486 https://youtu.be/cs5cbxDClbM?si=IQIFAD38cVzLCs55&t=486
Basically if you have the actual "factual" information, use it directly instead of hoping the LLM will accurately extract it and use it as part of a function call. In this case they already know what the accurate URLs are, just use it.
- dartos 2y agoThe LLM _might_ do what you want. Where I currently work, our function calls regularly fail only to succeed flawlessly on a retry. (I believe we’re on the order of 10s of millions openai calls a day) These are non-deterministic systems. I wouldn’t even trust them to accurately extract text until you did a beam search or something similar to kind of average out different LLM outputs