Fetching the paper…

Expediting and Elevating Large Language Model Reasoning via Hidden Chain-of-Thought Decoding · Around