1 paper
Dongxin Guo, Jikun Wu, Siu Ming Yiu
Large reasoning models such as DeepSeek-R1 and OpenAI o1 generate extended chains of thought spanning thousands of tokens, yet their integration with retrieval-augmented generation…