I have been trying to separate the truth from the hype, and learn more about how LLMs work, and this explanation has been one of the best one I’ve read on the topic. You strike a very good balance by going deep enough, but still keeping it understandable.
A question: I remember using Wolfram Alpha a lot back in university 15+ years ago. From a user perspective, it seems very similar to LLMs, but it was very accurate with math. From this, I take that modern LLMs are not the evolution of that model, but WA still appeared to be ahead of it’s time. What is/was the difference?
Thanks, I almost didn’t post because it was an essay of a comment lol, glad you found it insightful
As for Wolfram Alpha, I’m definitely not an expert but I’d guess the reason it was good at math was that it would simply translate your problem from natural language into commands that could be sent to a math engine that would do the actual calculation.
So basically act like a language translator but for typed out math to a programming language for some advanced calculation program (like wolfram Mathematica)
Again, this is just speculation because I’m a bit too tired to look into it rn, but it seems plausible since we had basic language translators online back then (I think…) and I’d imagine parsing written math is probably easier than natural language translation
Wow, what an insightful answer.
I have been trying to separate the truth from the hype, and learn more about how LLMs work, and this explanation has been one of the best one I’ve read on the topic. You strike a very good balance by going deep enough, but still keeping it understandable.
A question: I remember using Wolfram Alpha a lot back in university 15+ years ago. From a user perspective, it seems very similar to LLMs, but it was very accurate with math. From this, I take that modern LLMs are not the evolution of that model, but WA still appeared to be ahead of it’s time. What is/was the difference?
Thanks, I almost didn’t post because it was an essay of a comment lol, glad you found it insightful
As for Wolfram Alpha, I’m definitely not an expert but I’d guess the reason it was good at math was that it would simply translate your problem from natural language into commands that could be sent to a math engine that would do the actual calculation.
So basically act like a language translator but for typed out math to a programming language for some advanced calculation program (like wolfram Mathematica)
Again, this is just speculation because I’m a bit too tired to look into it rn, but it seems plausible since we had basic language translators online back then (I think…) and I’d imagine parsing written math is probably easier than natural language translation