That's a pretty big "if". LLMs are by design entirely unlike GoFAI reasoning engines. It's also very debatable whether it makes any sense to try and hack LLMs i

That's a pretty big "if". LLMs are by design entirely unlike GoFAI reasoning engines. It's also very debatable whether it makes any sense to try and hack LLMs into reasoning engines when you could just... use a reasoning engine. Or have the LLM to defer to one, which would play to their strength as translators.