Re:LLMs use a surprisingly simple mechanism to retrieve some...
by tel on Monday April 01, 2024 @09:28PM
(#39901408)
I think part of what I suspect is going on here too is more computation and finiteness. It seems correct that LLM architectures cannot perform too much computation (unless you unroll it in the context).
On the other hand you can look at statistical model identification in, say, nonlinear control. This can absolutely lead to unboundedly long predictions.
it is always trivial to take one of these models and expose it's failure to operate semantically, but these cases are never in the marketing material.
Consider an associative model of addition, all numbers from -1bn to 1bn, broken down into their digits, so that 1bn = <1, 0, 0, 0, 0, 0, 0, 0, 0>
Using such a model you can get the right answers for more additions than just -1bn to 1bn, but you can also easily find cases where the addition would fail.
It's never adding.