Search

10 results on this page

New Benchmark Tests Whether AI Can Draw Geometry

New Benchmark Tests Whether AI Can Draw Geometry

Researchers introduce an open-source benchmark separating olympiad geometry solving from accurate diagram construction. Across 954 problems, current foundation models achieved only a 36.14% average diagram compilation success rate, revealing a substantial gap between mathematical reasoning and faithful visual construction.

ArXiv AIResearch22h ago#geometry-reasoning#benchmark#asymptote
🎬

Kling AI Spins Out at $18B Valuation

Kuaishou is spinning out its Kling AI video-generation model into a separately financed company, attracting investment from Tencent, Alibaba Cloud, Baidu, and other institutions at an implied valuation of $18 billion. Kling has surpassed 100 million global users and nearly 50,000 enterprise customers, but continues to generate substantial losses because of compute costs.

Page 2