DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning
GRPOGRPO9 is the RL algorithm that we use to train DeepSeek-R1-Zero and DeepSeek-R1. It was originally proposed to simplify the...
DeepSeekR1
GRPOGRPO9 is the RL algorithm that we use to train DeepSeek-R1-Zero and DeepSeek-R1. It was originally proposed to simplify the...
