I mean the poster above you is wrong, they use math tools internally now when you ask math questions. Very obvious in Gemini. Yes the raw LLM trying to autocomplete the answer to a math problem is gonna be wrong but that’s not the way they are used to solve problems like that anymore.
I mean the poster above you is wrong, they use math tools internally now when you ask math questions. Very obvious in Gemini. Yes the raw LLM trying to autocomplete the answer to a math problem is gonna be wrong but that’s not the way they are used to solve problems like that anymore.
no way i’d want to drive on a bridge built on their supposed math
The LLM has to choose to use the calculating tools. Gemini tried to do this one solo:
Tbf, it did four of these calculations, and 75% were correct.
That makes sense. I clearly don’t keep up on the frontier models…