News for hackers. Stuff that matters.

A comment on How RLHF Works. Parent comment

Re:How RLHF Works by trade_monkey on Wednesday June 21, 2023 @04:27PM (#36423929)
RLHF is a band aid on not having enough data that fits your own biases and answers you want the model to give.
[ Parent | on HN ]
Re:How RLHF Works by astrange on Wednesday June 21, 2023 @09:18PM (#36426949)
It won't give answers at all if you don't train it to. It will output more questions because that's a more obvious completion to an incoming question.
[ Parent | on HN ]