> Those rigorous validation steps were also created with LLMs, correct?
Not really. I do explain in the methodology post how good o1-pro is at the task, but there was a lot of manual effort involved in coming to that conclusion with my own effort to review the LLM's reasoning, and even still, o1-pro is not perfect.
Not really. I do explain in the methodology post how good o1-pro is at the task, but there was a lot of manual effort involved in coming to that conclusion with my own effort to review the LLM's reasoning, and even still, o1-pro is not perfect.