1 paper
Kexin Sun, Yucong Guan, Jiaqi Sun +6
LLMs have shown strong potential for automating code review, yet their practical utility depends heavily on the design of generation and context strategies. In this paper, we inves…