If I'm reading this correctly, the author is saying that it's not a failure of logical deduction if the training data doesn't include the reversal. In other words, he's saying that if the data contains "Tom Cruise is the son of Mary Lee Pfeiffer” but not “Tom Cruise’s mother is Mary Lee Pfeiffer”, then the model's inability to determine the latter is "an explanation of how neural networks function than a model’s inability to deduce B is A."
But of course "how neural networks function" is that they fail at basic logical deduction and do not generalize.
So again, if I'm reading it correctly, he's hand-waving away the inability to make basic logical deductions because that not something they can or should be expected to do. As I read it, that means the reversal curse only exists if the answer to the question "can LLMs do logical deduction?" is "yes". If one takes the position that LLMs can't do general logical deduction, which seems to be the author's point of view, then there's no expectation that knowing "Tom Cruise is the son of Mary Lee Pfeiffer" is sufficient to determine “Tom Cruise’s mother is Mary Lee Pfeiffer”.
OK, but how does that not just demonstrate LLMs can't generalize absent massaged training data?
> it's not as stable as you'd want.
Which I take to mean that a model can sometimes confabulate "B is A", solely out of random variation, and that it's possible to bias the data and prompt to generate the expected response. The model hasn't done any logical deduction, the response is just a bias-influenced lucky break.
But of course "how neural networks function" is that they fail at basic logical deduction and do not generalize.
So again, if I'm reading it correctly, he's hand-waving away the inability to make basic logical deductions because that not something they can or should be expected to do. As I read it, that means the reversal curse only exists if the answer to the question "can LLMs do logical deduction?" is "yes". If one takes the position that LLMs can't do general logical deduction, which seems to be the author's point of view, then there's no expectation that knowing "Tom Cruise is the son of Mary Lee Pfeiffer" is sufficient to determine “Tom Cruise’s mother is Mary Lee Pfeiffer”.
Am I missing something?