Soonho Kong

DeRL: Diverse‑Exploration Reinforcement Learning for Large Language Models Improves Mathematical Reasoning

Synthetic Theorem Generation in Lean