Let it Calm: Exploratory Annealed Decoding for Verifiable Reinforcement Learning Paper • 2510.05251 • Published Oct 6 • 7