[cateye]

> we are inspired by the recent advancements in reinforcement learning (e.g., o1)

It is interesting to see what the future will bring when models incorporate chain of thought approaches and whether o1 will get outperformed by open source models.