Unverified50% confidenceFactExact time
GPT-3之后研究者发现通过精心设计的提示，单一大模型即可零样本或少样本完成多类任务，这一范式被称为In-Context Learning
1
Sources
50%
Confidence
Long-term
Relevance
7/18/2026
First Seen
Sources
Related Claims
VerifiedThe in-context learning capability of Few-Shot prompting was first systematically described in the GPT-3 paper.84% similarUnverifiedGPT系列采用单向（因果）注意力掩码，使模型在预训练阶段每个token只能关注其之前的上下文，与BERT等双向模型形成根本区别75% similarUnverified谷歌DeepMind的研究发现GPT类模型在特定提示条件下可精确复现训练集中约1%的文本片段75% similarUnverifiedSelf-supervised learning is the training foundation of the GPT series of models, which learn by predicting the next masked word in text.74% similarUnverifiedGPT-4o and subsequent models make sustained focus of 30 minutes or more possible through longer context windows, reinforcement learning alignment optimization, and improved chain-of-thought mechanisms.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/550183API
curl https://kongchang.com/api/v1/knowledge/claims/550183MCP
get_claim(id=550183)