The second batch of “First Proof” problems is meant to evaluate AI’s usefulness for research-level math. The best model got six or seven of the ten questions right.
These implementations are for demonstration purposes. They are less efficient than the implementations in the Python standard library.
Evaluating Research-Level Math Proofs via Strict Step-Level Verification Yifeng Sun Test-time Adversarial Takeover: A Real-time Hijacking Interface against Robotic Diffusion Policies Zi Yin , Peilin ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results