logoalt Hacker News

danielmarkbruceyesterday at 4:53 PM1 replyview on HN

RLVR and RLCR really don't need a whole bunch of special data.


Replies

verdvermyesterday at 6:34 PM

the algorithms technically, sure, however the outcomes definitely depend on data quality and coverage like any other training method, this is well known

show 1 reply