Primary sourceNoam Brown (OpenAI 推理)
View sourceEvent ·Single source
Noam Shazeer感慨:2.5年前实验室还在炫耀GSM8K成绩
Polymath AI 的 Noam Shazeer 发帖回顾,2.5 年前各 AI 实验室还会用小学数学基准 GSM8K 上的成绩来宣传自家前沿 LLM。GSM8K 曾是衡量模型推理能力的标志性基准,如今已被更强的数学与推理评测取代。他表示期待再过 2.5 年模型能力会到什么程度。
Current read
Shazeer 发推感慨:2.5 年前大家还拿小学数学题 GSM8K 吹模型,现在早不够看了,两条对比很直观。
2 sources33° AI heatLast updated 2026/10/27 00:41:00
Evidence
2 sourcesRelated source 1arXiv cs.LG
View sourceConflict check
Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.