News for hackers. Stuff that matters.

A comment on An analysis of DeepSeek's R1-Zero and R1. Parent comment

Re:An analysis of DeepSeek's R1-Zero and R1 by godelski on Thursday January 30, 2025 @02:15AM (#42875649) 

  > I highly doubt you are getting novel, high quality data.
That's not the point. The point is you reject low quality data, aka noise
[ Parent | on HN ]
Re:An analysis of DeepSeek's R1-Zero and R1 by hassleblad23 on Thursday January 30, 2025 @04:02AM (#42876178)
And how would that work at inference time?
[ Parent | on HN ]
Re:An analysis of DeepSeek's R1-Zero and R1 by bobxmax on Thursday January 30, 2025 @08:26AM (#42877468)
Why would it need to work at inference time?
[ Parent | on HN ]