Chase Blagden

Just Enough Thinking: Efficient Reasoning with Adaptive Length Penalties Reinforcement Learning

Self-Guided Thinking: Enabling LLMs to Decide When to Think

Generative Reward Models