Apr 20, 2024 — 4 min read What is the mathematical formulation of reasoning? How can we make LLMs like chatGPT think before they speak? And how can we make that baked into the model so it can learn to think in a self-supervised way without having to "explain it step by step" (or another famous prompt we use when we want to improve chatGPT performance drastically)? How can we teach models to think ahead? I will share with you the results of some experiments that may cast light on the path of "Rea...