← Back to context

Comment by bojan

5 hours ago

I agree those results ate handy, but I had several occasions where they turned out to be completely wrong. 95% correctness rate is not good enough.

LLMs have a lot of issues with facts, because they are probabilistic and you typically only get one answer per query instead of multiple covering a larger space.

However, they are still useful in these cases if you know the above and use their output as a starting point to think and ask questions.