Open Lab 17 · Reinforcement Learning from Verifiable Rewards