News for hackers. Stuff that matters.
Re:AI is going to eat itself: Experiment shows people traini... by raverbashing on Friday June 16, 2023 @10:31AM (#36357130)
I'm very not sure about this

Text quality in reddit is awful. Especially comments . Forget using it for any semblance of actual grammatic correctness.

Probably can be used, after heavy cleaning and curation.

Re:AI is going to eat itself: Experiment shows people traini... by Dr_Birdbrain on Friday June 16, 2023 @10:39AM (#36357272)
I think in this context, grammatical correctness is irrelevant, and what you actually want is just raw text as written by real humans from all walks of life.
Re:AI is going to eat itself: Experiment shows people traini... by juujian on Friday June 16, 2023 @10:54AM (#36357577)
Yup, variety is a big part of it. You don't just want a large language model that can produce meticulous text, you also need to be able to handle user input, which will frequently include faulty input itself. Otherwise you end up with something as mechanic and inflexible as google assist. Interactions with large language models can be much more organic if you have this kind of input data.
Re:AI is going to eat itself: Experiment shows people traini... by ben_w on Friday June 16, 2023 @11:19AM (#36358078)
Agreed. We can make infinite grammatically correct text with perfect spelling, but if we limit the training set to that alone then ppl who wright liek this cant uze the ai lol & will be all lije "wtf m8 this aint skool i don care bout spekling rite u know wot i ment y cant ur machine?!?"

…

and while I enjoy my ivory tower as much as the rest of us, I do think it's important for the AI to understand everyone, not just me.

Re:AI is going to eat itself: Experiment shows people traini... by MattGaiser on Friday June 16, 2023 @10:54AM (#36357579)
Depends on what you are looking to create. Good writing? Then it is an issue. Authentic seeming writing? Then it is exactly what you want.