Is it accurate to compare 8k example RL with 8k example SFT? RL with the same amount of examples would take massively more compute than the SFT version (though depending on how many rollouts they do per example).
RL is more data-efficient but that may not be relevant now that we can just use Deepseek-R1's responses as the training data.
RL is more data-efficient but that may not be relevant now that we can just use Deepseek-R1's responses as the training data.