RunByAI

Tag: vector database

2 post(s)

September 20, 20260 comment(s)

Introduction to Rerank: Why do we have to "re rank" after RAG retrieval

When the Retrieval Enhanced Generative (RAG) system was first launched, many people felt that the answer was "irrelevant": the knowledge base clearly had the correct paragraph, but the model reference

Tech Guide Search enhanced generation vector database RAG

September 12, 20260 comment(s)

Vector Databases 101: How to Choose and Tune the Retrieval Layer for RAG

检索增强生成（RAG）已经成为大模型落地最主流的工程范式之一：它不改动模型权重，而是把外部知识以“检索 + 拼接”的方式送进上下文。问题也随之而来——很多团队把精力全花在提示词和模型选型上，却低估了检索层的重要性。检索没召回对，再强的模型也只能基于错误信息作答。一、向量数据库到底解决什么问题RAG 的核心是把文本转成向量（embedding），再按语义相似度找出最相关的片段。向量数据库就是承载这一

Tech Guide RAG vector database