AI models are starting to crack high-level math problems AI models are starting to crack high-level math problemsÂ
RiseStock on January 16, 2026 2:23 pm I don’t think the reporting is accurate. The models are not themselves proving theorems. The models are usually paired with lean or other proof languages and iteratively changing the outputs until something valid comes out.
5 Comments
Sure. Let’s see the (human) peer reviews.
I don’t think the reporting is accurate. The models are not themselves proving theorems. The models are usually paired with lean or other proof languages and iteratively changing the outputs until something valid comes out.
Clickbait
Yeah ok
Adding machine good at adding… Film at 11.