#Grpo

...or how to make Qwen learn query optimization via agentic reinforcement learning

RRCM uses GRPO to learn when to retrieve evidence for LLM recommendation, outperforming fixed-context baselines.