I think it's more RLVR (reinforcement learning from verified rewards). The RLHF ... | Hacker News

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

visarga 18 days ago | parent | context | favorite | on: GPT-5.6

I think it's more RLVR (reinforcement learning from verified rewards). The RLHF is just to align models to human preferences, meaning to behave nice.

versteegen 18 days ago | [–]

More accurate to say RLHF aligns models to human preferences, most significantly to be helpful.

redanddead 18 days ago | [–]

What makes you say that

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