新 SOOHAK 基准测试暴露 AI 数学缺陷
名为 soohak 的新基准测试显示，前沿模型在研究级数学上得分低于 30%，且无法识别无解问题。
developed by carnegie mellon, eleutherai, and seoul national university, soohak features 340 research-level problems and 99 unsolvable 'refusal' tasks to test calibration. while gemini 3 pro achieved 61 percent coverage on a subset compared to human mathematicians' 51 percent, the benchmark reveals that increased compute improves solving but not the ability to admit a problem has no answer.