1 paper · 1 filter
Zheyang Xiong, Shivam Garg, Max Yu +4
Long Chain-of-Thought (CoT) reasoning improves LLM problem-solving but is computationally expensive due to sequential token generation. While recent works explore reasoning in cont…