deepseek-r1 incentivizing reasoning capability in llms via reinforcement learning

quickq官方安卓版