News for hackers. Stuff that matters.
Re:"Hallucinating" AIs sound creative, but let's not celebra... by pdonis on Saturday October 14, 2023 @02:16PM (#37882839)
> Conceptually this is the same for LLMs.

No, it isn't. LLMs don't look at the results of their actions and update a world model. Humans do.

Re:"Hallucinating" AIs sound creative, but let's not celebra... by nuancebydefault on Saturday October 14, 2023 @04:41PM (#37884081)
This is indeed a difference that I overlooked, there is limited feedback on LLMs output, other than a rater primitive survival model (models that aren't popular/correct/cheap/fast enough are not used anymore over time)