News for hackers. Stuff that matters.

A comment on Reinforcement Pre-Training.

Re:Reinforcement Pre-Training by dgshsg on Tuesday June 10, 2025 @09:06AM (#44236326)
I notice that you can do this recursively to arbitrary depth. The cost is terrible though.
[ Parent | on HN ]
Re:Reinforcement Pre-Training by watsonmusic on Tuesday June 10, 2025 @12:37PM (#44238797)
it could be adaptive. only high-value tokens were allocated with more compute
[ Parent | on HN ]