SolidGeo is the first large-scale benchmark specifically designed to evaluate the performance of MLLMs on mathematical reasoning tasks in solid geometry. SolidGeo consists of 3,113 real-world K–12 and ...

Google’s AlphaGeometry2 AI system, developed by DeepMind, solves geometry problems at a level beyond top human competitors. Credit: Alpha Photo / CC BY-NC 2.0 Google scientists have developed a ...