Does optimizing context actually make a reasoning model think less?

Introduction LLMs have become very popular, with one big use case being people asking questions by just providing the context along with them. In this article, we analyze how Gemini's frontier reasoni