How You Begin is How You Reason: Driving Exploration in RLVR via Prefix-Tuned Priors

arXiv:2605.08817v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) recently thrives in large language model (LLM) reasoning tasks. However, the reward sparsity and the long reasoning horizon make effective exploration challenging. In practice, this challenge...