Unverified50% confidenceFactExact time

O1模型的性能可沿训练时算力（Train-Time Compute）和推理时算力（Inference-Time Compute）两个维度持续提升

1

Sources

50%

Confidence

Medium-term (~90 days)

Relevance

7/2/2026

First Seen

Valid until: 9/30/2026

Sources

Andrew Ng's New Course Explained: A Practical Guide to Using OpenAI's O1 Reasoning Model

bilibili吴恩达智能体6/13/2026

Related Claims

Verified推理时扩展（inference-time scaling）通过在推理阶段增加计算量提升模型性能，存在类似预训练的幂律关系76% similar Unverifiedo1系列模型的核心机制是推理时计算扩展（inference-time compute scaling），通过在推理阶段生成长链式思考过程来进行深度推理76% similar Verified推理时计算扩展（Test-time Compute Scaling）通过在推理阶段投入更多计算资源来提升模型性能，研究表明在某些任务上比单纯增大模型参数更加高效74% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/47315

API

curl https://kongchang.com/api/v1/knowledge/claims/47315

MCP

get_claim(id=47315)