That's basically what i was saying. It's close to human capability. I don't really know how to explain it, but to me we have at least two kind of reasoning, sw

That's basically what i was saying. It's close to human capability. I don't really know how to explain it, but to me we have at least two kind of reasoning, swallow and deep. LLMs could well be capable of the first.