Comment by djhn

3 hours ago

Research seems to on balance point towards RLHF&RLVR merely increasing subjective sampling efficiency within the pretraining data.

0 comments