The Inefficiency of Static Chain-of-Thought Reasoning in LRMs Recent LRMs achieve top performance by using detailed CoT reasoning to solve complex tasks. However, many simple tasks they handle could ...