Comment by qwery
14 days ago
Why would that thrill linguists? I'm not saying it hasn't/wouldn't/shouldn't, but I don't see why this technology would have the dramatic impact you imagine.
Is/was the same true for ASCII/Smalltalk/binary? They are all another way to translate language into something the computer "understands".
Perhaps the fact that it hasn't would lead some to question the validity of their claims. When a scientist makes a claim about how something works, it's expected that they prove it.
If the technology is as you say, show us.
> Is/was the same true for ASCII/Smalltalk/binary? They are all another way to translate language into something the computer "understands".
That's converting characters into a digital representation. "A" is represented as 01000001. The tokenization process for an LLM is similar, but it's only the first step.
An LLM isn't just mapping a word to a number, you're taking the entire sentence, considering the position of the words and converting it into vectors within a 1,000+ dimensional space. Machine learning has encoded some "meaning" within these dimensions that goes far far beyond something like an ASCII string.
And the proof here is that the method actual works, that's why we have LLMs.