snell2024-test-time-compute

arXiv： 2408.03314

TLDR（中文）

系统性给出"推理时多花 compute"的 scaling law：在固定预算下，对小模型加推理时搜索往往比训练更大模型更划算。是 o1/R1 时代理论支撑。

TLDR (English)

Systematically presents scaling laws for "spending more compute at inference time": with fixed budget, adding inference-time search to small models often more cost-effective than training larger models. Theoretical foundation for o1/R1 era.