HN Explorer

5 comments

[ 3.5 ms ] story [ 24.0 ms ] thread
ekadagami
Is this something similar to a reranker, like JEV, for example?
senorcarbone
I think they simply use the LLMs softmax scores (uncalibrated confidence) so the answer is yes
senorcarbone
From hallucinated output to hallucinating decisions :). The only way to automate decisions reliably is to do so without AI in the middle but just trustworthy statistics. Skeptical does this right https://skeptical.io
ziranbing
Artificial Intelligence belongs to the future