Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

My takeaway from this kind of thing is that it demonstrates what OpenAI has accomplished as far as managing to be so much "smarter" on average, or even when compared to what Google has available.

My guess is that the amount of RLHF they have done is much more than any other project.



You are aware RLHF tends to make models dumber, though, right?




Consider applying for YC's Winter 2027 batch! Applications are open till November 2.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: