1 paper
Lianlei Shan, Han Chen, Yixuan Wang +2
While Large Language Models (LLMs) demonstrate exceptional performance in surface-level text generation, their nature in handling complex multi-step reasoning tasks often remains o…