News for hackers. Stuff that matters.
Re:91% of ML Models Degrade over Time by barrkel on Saturday April 15, 2023 @03:04AM (#35578319)
You're presenting an example of inference of rules from given data as a counterexample for novelty. They're not even in the same category of thing. Invention is not learning. Sometimes invention is interpolation, but sometimes it isn't: corridors is an interesting example, because they are not obviously a remix of anything.
Re:91% of ML Models Degrade over Time by fenomas on Saturday April 15, 2023 @05:41AM (#35579058)
No, I presented it as a counterexample to the claim that LLMs only statistically regurgitate existing data.