TechCrunch · AI· Tim Fernholz·· 8 d ago
OpenAI’s math solutions aren’t meeting the field’s standards yet
OpenAI’s math solutions aren’t meeting the field’s standards yet
AI summary
OpenAI's flood of proofs deviated from the guidelines set by a group of mathematical researchers consulted by the frontier lab.
Selection record
Threshold 76Media and individualsFirst 62Second 62
Not admittedSum of both 124 < twice the threshold 152
- Source tier
- Media and individuals; this tier's threshold is 76
- Pre-filter
- passed：OpenAI模型解数学题及AI形式化证明争议
A model scores each item twice, independently, against one written standard, out of 100. An item is admitted only when the two scores add up to twice the threshold. The threshold is set per source tier.
Source: TechCrunch · AI · techcrunch.com