热点
"COUNTERMATH" 相关文章
ICML 2025 | 会做题≠会思考？首个反例驱动推理基准：揭穿大模型“刷题式假象”
PaperWeekly 2025-08-29T13:19:17.000000Z
Boosting AI Math Skills: How Counterexample-Driven Reasoning is Transforming Large Language Models
MarkTechPost@AI 2025-02-21T04:59:47.000000Z