Says in the papers. "...which was first mathematically generated in one shot by an internal model at OpenAI, and then expositionally refined through
human interactions with Codex."
Doesn't really matter the prep-work, what they say is it's a one-shot result, achieved by AI. The blog doesn't claim it was done by a currently public Model.
The model doing it one shot does not mean they only attempted it once though. They could have tried and retried a ‘one-shot’ answer hundreds of times before it produced a workable result
Doesn't really matter the prep-work, what they say is it's a one-shot result, achieved by AI. The blog doesn't claim it was done by a currently public Model.