向量 / 嵌入Embedding / Vector
Embedding / Vector
解决什么问题What problem it solves
把一段文字变成一串数字。机器于是能按「意思相近」而不是「字面相同」来检索。Represent text as coordinates, so meaning can be compared by geometry rather than by string match.
这是什么What it is
把文本映射成高维数字数组，相近含义的文本会得到相近的向量。
它是「语义搜索」的实现基础。搜「怎么退货」能命中讲「退款流程」的文档，靠的是这个。
Turning text into a list of numbers where closer vectors mean closer meaning. It is how semantic search and retrieval work.
为什么需要知道Why it matters
它解释了 RAG 里的「检索」到底在检索什么。它也是向量数据库存在的原因。Most retrieval systems are built on it. It also explains why keyword search still matters alongside vector search.
现在能做什么What you can do now
向量与上下文Vectors and context (context7) ↗
回到名词地图看第2站的其他 35 个词 →Back to the map: the other 35 terms in station 2 →
出处Sources
- 向量与上下文Vectors and context (context7) ↗