News for hackers. Stuff that matters.
Re:An OpenAI model has disproved a central conjecture in dis... by slashdave on Wednesday May 20, 2026 @08:24PM (#48216205)
You have to define what you mean by "interpolate". The mechanisms that LLM use are not mysterious, and they are not the same as used by humans.
Re:An OpenAI model has disproved a central conjecture in dis... by drdeca on Wednesday May 20, 2026 @09:11PM (#48216537)
If you interpret “interpolate” in the literal sense, and apply it to the mechanisms behind LLMs, then the claim that they only interpolate, is straightforwardly false.

Taking it instead as a metaphorical claim may be more valid, but in that case it doesn’t depend on our understanding of how LLMs work.

Re:An OpenAI model has disproved a central conjecture in dis... by slashdave on Wednesday May 20, 2026 @10:53PM (#48217263)
LLMs are statistical models by construction, so depending on how liberal you want to be with terminology, "interpolate" is not so bad. Might make a statistician upset.
Re:An OpenAI model has disproved a central conjecture in dis... by baq on Thursday May 21, 2026 @01:42AM (#48218349)
They’re also capable of performing arbitrary computation when ran in a loop - so they can be made to quite literally interpolate whatever. Philosophers are quite upset, too.
Re:An OpenAI model has disproved a central conjecture in dis... by slashdave on Sunday May 24, 2026 @01:59AM (#48254816)
Define "arbitrary". Without RAG, they screw up basic algebra.
Re:An OpenAI model has disproved a central conjecture in dis... by drdeca on Thursday May 21, 2026 @02:07AM (#48218540)
But people aren’t giving a (less literal) definition of what they mean by “interpolate” that relies on the internal mechanisms of these models, just a vague metaphor, which, as this vague metaphor, there’s nothing it uses about LLMs that makes the question “do LLMs just interpolate” less of a type error than “do people just interpolate”.

And I don’t think it’s a good metaphor.