News for hackers. Stuff that matters.

A comment on Mistral NeMo. Parent comment

Re:Mistral NeMo by michaelt on Thursday July 18, 2024 @12:50PM (#40997458)
I don't have explanations but I can point you to one of the papers: https://arxiv.org/pdf/2309.12288 which calls it "the reversal curse" and does a bunch of experiments showing models that are successful at questions like "Who is Tom Cruise’s mother?" (Mary Lee Pfeiffer) will not be equally successful at answering "Who is Mary Lee Pfeiffer’s son?"
[ Parent | on HN ]
Re:Mistral NeMo by spookie on Thursday July 18, 2024 @02:54PM (#40998623)
Isn't that specific case just a matter of not having enough data _explicitly_ stating the reverse? Seems as if they are indeed stochastic parrots from that perspective.
[ Parent | on HN ]
Re:Mistral NeMo by layer8 on Thursday July 18, 2024 @05:31PM (#40999964)
Yes, and your conclusion is correct.
[ Parent | on HN ]
Re:Mistral NeMo by mlyle on Friday July 19, 2024 @12:07PM (#41007904)
You know, I'm not sure that humans are so good at that kind of reverse task. Information can be very easy to access from one direction but very hard to reach from others. We're not databases.
[ Parent | on HN ]