Unverified50% confidenceFactExact time
嵌入思想源自2013年Word2Vec提出的分布式语义假说，即词的含义由其上下文分布决定
1
Sources
50%
Confidence
Long-term
Relevance
7/23/2026
First Seen
Sources
AI构建者手册：LLM原理、RAG架构与智能体设计全解
redditr/aiagents7/12/2026
Related Claims
UnverifiedThe idea of embedding vectors can be traced back to Word2Vec proposed by Mikolov et al. in 2013, whose core assumption is the distributional hypothesis.78% similarUnverified分布假说指出意义相近的词出现在相似的上下文中，导致反义词在Word2Vec、GloVe、BERT等模型中学到相近的表征方向73% similarUnverified向量嵌入的核心思想源于分布式语义假说，即语义相近的词或句子在向量空间中距离更近68% similarVerifiedThe essence of large language models is learning the conditional probability distribution of what the next word is given the preceding text64% similarUnverified大语言模型是概率性、随机性的，它根据之前出现的词元来预测下一个词元64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/599977API
curl https://kongchang.com/api/v1/knowledge/claims/599977MCP
get_claim(id=599977)