Bingo: Boosting Efficient Reasoning of LLMs via Dynamic and Significance-based Reinforcement Learning

Open in new window