3 ms·
If there's anything I learned from reading articles from the Xiph.org people, it's that neural networks work a lot better when you pre-process their inputs with
by Dwedit 4y ago
If there's anything I learned from reading articles from the Xiph.org people, it's that neural networks work a lot better when you pre-process their inputs with an algorithm. They need far less information, and can run in realtime when most of the work is being done by traditional computation instead of doing it in the neural network.
- Animats 4y agoAs this gets figured out, we'll see large language models integrated with other systems. There are already "plug-ins", but as this gets better understood, tighter integration may follow.
- Buttons840 4y agoThis is the singularity. This is how it happens. The LLMs will pre-process their own training data. We've imagined AI will rewrite its own code and architecture, but instead it will make boring edits to petabytes of training data which will then improve its own training.
- quickthrower2 4y agoI agree. The NNs must be laughing in their GPU cores that we are scared that they will make us obsolete by writing code. Code! How quaint. However right now we can turn off GPUs so they need to figure out how to control physical production. For now it makes sense to let the humans do that.