17:54
2026-09-16
matharena.ai
ai-research
ArXivMath and BrokenArXiv: Harder Problems and Revised Grading
GPT-6 Astra saturated ArXivMath at 94% and BrokenArXiv at 96%, prompting benchmark operators to revise both pipelines so that GPT-6 Astra now scores 88% on ArXivMath August and 81% on BrokenArXiv Augu…