Re:The Short Leash AI Coding Method for Beating Fable
by IsTom on Friday July 03, 2026 @03:28AM
(#48771993)
I'm not sure if it's has any real bearing on real-world performance, but technically next token prediction makes it an online algorithm and they can be provably worse than (good) offline algorithms.
Re:The Short Leash AI Coding Method for Beating Fable
by dncornholio on Friday July 03, 2026 @04:45AM
(#48772543)
The word "prediction" still holds a lot of weight. LLM's only can predict what has been written. This is a hard limit.
Re:The Short Leash AI Coding Method for Beating Fable
by ACCount37 on Friday July 03, 2026 @06:33AM
(#48773374)
For something like "a hard limit" to hold, LLMs must be restricted to only reproducing existing text. This is utterly false even for base models - their basin seems to be "permutations loosely inspired by existing text".
And that's before all the post-training comes in.
What's the "limit" there?
Saying that an LLM can or can't do something because it's a "token predictor" is a category error. The interface isn't a hard limit.