UR2: Unify RAG and Reasoning through Reinforcement Learning.

Weitao Li, Boran Xiang, Xiaolong Wang, Zhinan Gou, Weizhi Ma, Yang Liu