Comment by vagab0nd

5 months ago

Cool trick. But is this better than reinforcement learning, where the LLM decides for itself the optimal thinking time for each prompt?

0 comments