Breaking the Exploration Bottleneck: Rubric-Scaffolded Reinforcement Learning for General LLM Reasoning Paper • 2508.16949 • Published Aug 23 • 22