I have what is probably a stupid and misplaced question. The second picture in the article has the phrase “with hope in his heart”. That phrase repeatedly pops up in the hilariously bad ChatGPT stories I’ve seen people generate.
Is there a reason that cheesy phrases that don’t get used in real life keep popping into stories like that?
I want to preface this by saying I’m not doubting you, I just don’t know how it works.
Ok, but wouldn’t the training be weighted against older phrases that are no longer used? Or is all training data given equal weight?
Additionally, if the goal is to create bedtime stories or similar, couldn’t the person generating it ask for a more contemporary style? Would that affect the use of that phrase and similar cheesy lines that keep appearing?
I would never use an LLM for creative or factual work, but I use them all the time for code scaffolding, summarization, and rubber ducking. I’m super interested and just don’t understand why they do the things they do.
That’s not the point of this I think lol. It’s very impressive as a tech demo, that even a device even as underpowered as a Pi can run these AI models to a passable degree.
That’s what I meant. Ai stories are not passable and I think I think if we give them to people who don’t know how stories work (children) we are in for a bad time
The article does say it takes five minutes to create a new story and picture. I assume most of that time is spent generating the picture. Still pretty impressive, but nowhere near the few seconds you can get with fast hardware.