It will be interesting to see if it can come up with a cheaper to construct graviton detection experiment
replyMost of high energy theoretical physics is very non-rigorous or even hand-wavy.
I think AI isn’t there yet for such problems.
replyPlease make a benchmark for it, that'd be super interesting! My guess is we'd see models climbing it quickly, but maybe not
reply