If you gave me a model released two years ago and today and let me do some programming with both, I would have no problems telling you which one was released two years ago; progress on this front is very noticeable. But if you let me chat with each one for an hour, I'm honestly not sure I would be able to tell the difference.
Research into machine learning, transformers, there's tons of technicalities that enables LLMs. ANd there's more and more "technicalities" being thrown on top of LLMs and we're having modifications to LLMs as well.
LLMs are the tip of a spear of a trendline that didn't involve LLMs. Prior to that we had AI generating art and music through diffusion algorithms. We had AI doing image recognition and doing mind reading. The trendline is clear to anyone but those who think the current state of LLMs and the problems we have with it are completely static in nature.
I cannot even begin to express how great this looks. I hit that again and again and .. it's a papercut, but so damn annoying. Thank you. Very much appreciated.
(Now, is there an addon to fix the tablet UI back to 'simple'?)
