News for hackers. Stuff that matters.

A comment on RL is more information inefficient than you thought. Parent comment

Re:RL is more information inefficient than you thought by ACCount37 on Sunday November 30, 2025 @01:42PM (#46099186)
RL is very important - because while it's inefficient, and sucks at creating entirely new behaviors or features in LLMs, it excels at bringing existing features together and tuning them to perform well.

It's a bit like LLM glue. The glue isn't the main material - but it's the one that holds it all together.

[ Parent | on HN ]
Re:RL is more information inefficient than you thought by elchananHaas on Sunday November 30, 2025 @10:49PM (#46103288)
RL before LLMs can very much learn new behaviors. Take a look at AlphaGo for that. It can also learn to drive in simulated environments. RL in LLMs is not learning the same way, so it can't create it's own behaviors.
[ Parent | on HN ]