Unverified50% confidenceFactExact time
Embedding models such as OpenAI's text-embedding-ada-002, BGE, and E5 map text into points in a high-dimensional vector space so that semantically similar texts are closer together.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026
Sources
Related Claims
UnverifiedOpenAI的text-embedding-ada-002能将文字映射到高维向量空间，语义相近的文本在向量空间中距离更近78% similarVerifiedOpenAI 的 text-embedding-3-small 模型会将文本转换为 1536 维的浮点数数组70% similarUnverified向量数据库通过嵌入模型（如OpenAI的text-embedding-3-large或开源的BGE-M3）将文本转化为数百至数千维的浮点数向量，向量距离对应语义相似度70% similarUnverified嵌入模型可将文本压缩为1536维或1024维的浮点向量，OpenAI的text-embedding-ada-002和Cohere的embed-multilingual是常见示例70% similarUnverified主流嵌入模型包括 OpenAI 的 text-embedding-ada-002 和 Sentence-BERT 系列，余弦相似度是最常用的相似性度量69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/44761API
curl https://kongchang.com/api/v1/knowledge/claims/44761MCP
get_claim(id=44761)