MathAdv: What Theorem Provers Know, Reason, Formalize, and Generalize
This work introduces MathAdv, a diagnostic benchmark spanning 13 domains across undergraduate- and graduate-level mathematics, and shows how component-wise evaluation can reveal model capabilities and failure modes that aggregate theorem-proving accuracy obscures.
Jiajie Yuan, Connor Martinez Lockhart, Xiao-Yun Liu et al.
· 0 citations