精选理由
数学基准更新，GPT-5.5和Google AI成绩亮眼
Epoch AI 发布 FrontierMath 基准测试 v2 版本，修复了 42% 的问题错误。新版本中，GPT-5.5 (xhigh) 在 Tier 1-3 上取得 85% 的准确率，Google 的 AI co-mathematician 在 Tier 4 上达到 76%。所有模型得分普遍提高，排名基本不变。
原文 · Epoch AI
FrontierMath: Tiers 1–4 (v2) is live. We conclude…
FrontierMath: Tiers 1–4 (v2) is live.
We concluded an audit that addressed errors in 42% of problems. Rankings are similar but scores are higher across the board. The current leaders are GPT-5.5 (xhigh) with 85% on Tiers 1–3 and Google’s AI co-mathematician with 76% on Tier 4. https://t.co/DH9nhpKH0N