It still can’t count the Rs in strawberry, I’m not worried.

[email protected]

I mean I tested it out, even tbough I am sure your trolling me and DeepSeek correctly counts the R's

[email protected]

Not trolling you at all:

https://lemmy.world/comment/14735060

[email protected]

Non thinking prediction models can't count the r's in strawberry due to the nature of tokenization.

However openai o1 and deep seek r1 can both reliably do it correctly

[email protected]

Yes it can

[email protected]

Screenshots please

https://lemmy.world/comment/14735060

[email protected]

[email protected]

https://lemmy.world/comment/14754440

[email protected]

It searches the internet for cats without tails and then generates an image from a summary of what it finds.

That's how this Machine Learning progam works

[email protected]

Clearly not the first try

[email protected]

“Again” so it failed the first time. Got it.

[email protected]

It didn't, I just wanted a short reply. Though it failed when I asked again at the same chat. But when asked to split the word to 2 parts it became sure that the correct answer is 3.

[email protected]

That’s a lot of processing just to count letters. Hopefully it can add numbers without splitting the number

[email protected]

so.... with all the supposed reasoning stuff they can do, and supposed "extrapolation of knowledge" they cannot figure out that a tail is part of a cat, and which part it is.

[email protected]

That isn't at all how something like a diffusion based model works actually.

[email protected]

So what training data does it use?

They found data to train it that isn't just the open internet?

[email protected]

The "reasoning" you are seeing is it finding human conversations online, and summerizing them

[email protected]

Regardless of training data, it isn't matching to anything it's found and squigglying shit up or whatever was implied. Diffusion models are trained to iteratively convert noise into an image based on text and the current iteration's features. This is why they take multiple runs and also they do that thing where the image generation sort of transforms over multiple steps from a decreasingly undifferentiated soup of shape and color. My point was that they aren't doing some search across the web, either externally or via internal storage of scraped training data, to "match" your prompt to something. They are iterating from a start of static noise through multiple passes to a "finished" image, where each pass's transformation of the image components is a complex and dynamic probabilistic function built from, but not directly mapping to in any way we'd consider it, the training data.