573 karma · joined June 13, 2023
The fact that this is possible also adds some pretty strong restriction to the set of possible maps you could use to remove that information. No linear map will work since all embedding spaces are ~an orthonormal matrix apart, so some form of encryption is necessary. This wasn't known until very recently.
I think you could get a lot closer by framing this as an optimization problem, where you use the full alphabet dictionary, but add a residual prediction which aims to cover as much of the remaining domain name tree as possible weighted by popularity. This tree could then be pre-baked and stored with the same system. This would probably get you p99 0ms even in Australia.
The choice function is also exactly the same as the pascal triangle!
Similar things are true of Bayesian stats, leading to things like predictively oriented posteriors being studied nowadays.
Edit: to be clear, this was random ones on YouTube, not sure about that site.
EDIT: Apparently not licensed, just completely unrelated.
On some level, I think the most straightforwardly creative work in maths comes from building up useful generalizations; I can't see current LLMs inventing even something as straight forward as a manifold from first principles; but I also don't think that's fundamentally required to prove useful things, I think it's more of a shortcut for humans.