Then they don't need to train on github, no? Why not release a new model trained from Knuth's Art of Programming, Cormen's Introduction to Algorithms and the C specification.
Feel free to throw in any other published literature related to STEM, but stick to the code samples from the books.
I'm certain it'll be able to change the color of a CSS button, right?
LLM can be trained on a code and at the same time reproduce the core ideas. That's what LLMs do after all - they convert the training data into their own internal models and representations, and then reproduce the ideas.
Sure, some things/patterns, that were repeated multiple times, LLMs will tend to repeat verbatim as well, but that's not that big of a problem.
As a person who invented a few algorithms on my own I absolutely love LLMs and I don't mind them being trained on my work, but yeah - I've been way less likely to publish open source over the last year. In the past, if some of my stuff got traction, the credit was close to automatic (early adopters credited or at least knew where they got it from). Nowadays, LLMs will train on these ideas, rewrite them, and give no credit.
Still, I prefer this to having no LLMs at all.
> but stick to the code samples from the books. > I'm certain it'll be able to change the color of a CSS button, right?
A good enough LLM will just decompile a browser, figure out CSS spec from it, and yes - figure out how to change the color of a CSS button from first principles. There is no point to do this with CSS, but with other things it's now easier to just dig through sorces or direct bytecode than to bother checking docs.
Can it decompile a browser using a specification of x86 and the source code of the compiler?
e.g. without training on the source code and binaries of all software it was able to rip from the internet?
Also, it would be relatively easy to build synthetic datasets for training.
I, for one, don’t mind models being trained on stuff I produced and shared publicly over the last 20 years. I did it for common good, including commercial uses, and this is one of them.
Plenty of people who never produced any open source trying to argue as if if they did.
Because they're really really stupid and only make up for this by being really really stupid really really fast.
This has been ruled, by actual courts, to not be "stealing" (not even in the "you wouldn't steal a car, piracy is theft" sense that film and music studios campaigned on).
The last I heard was the "Chinchilla" scaling law was ~20 training tokens per parameter. Humans are, if you'll excuse a very hand-waving Fermi estimate, 100,000 times more data-efficient at learning stuff (it's really hard to tell given we're visual creatures that happen to speak, while LLMs are text-based things that happen to see).
Just skip the source code. Consider it an easier challenge than reinventing relativity from 19th century physics.
This would be like teaching cooking without looking at any recipes, just from physics and first principles. Or learning music without looking at the sheet music / listening to any existing songs, just generic musical theory and chords. I don't think humans can do it "zero-shot" either...
And this implies that training on the "theoretical" side of things doesn't give you much insight on the "practical" side. Stuff like cyclomatic complexity, UML diagrams and all that stuff might be well-represented in literature but way less so in real software, so training on the literature will produce completely different software than training on production software code.
https://patentimages.storage.googleapis.com/db/8f/cb/dad63e9...
I'm going to stop arguing with you now.
What makes you think that wouldn't work? I think a lot of the hype around AI is vastly overblown but that seems to be well within the scope of what they can be expanded to do in the not too distant future. AlphaGo was trained through self-play reinforcement learning IIRC and I don't really see a reason that some sort of equivalent couldn't be done for generating code starting with textbooks and access to a Linux CLI as a reference. It would be an interesting experiment at least.