FLORIAN BASTIN
All sessions

Course 08

LLM reasoning

This session covers models trained to carry out longer reasoning on certain tasks.

We start from the methods seen in the previous session to see how reinforcement learning applies here. We also look at what these models cost and where they genuinely help.

On the programme
  1. 01Reasoning models
  2. 02RL for reasoning
  3. 03GRPO
  4. 04Scaling