docs(mvp): update RAG docs for py-rag extraction

- rewrite RAG architecture doc for py-rag service contract mapping, ingest and rebuild ops
- refresh observability/trace doc for post-L0 single-attempt semantics
- add py-rag chain exploration note; mark superseded Milvus/L0 notes with status banners
- close ISS-017 (superseded by L0 sinking); mark knowledge_domain table orphaned
- refresh architecture/mvp/engineering indexes
This commit is contained in:
zhuyongxin
2026-09-30 17:03:50 +08:00
parent 9cf162482d
commit 473bb5d004
16 changed files with 597 additions and 343 deletions
@@ -1,5 +1,10 @@
# 诊断 Agent 场景下的 RAG 排序:从 K=3 规则加分,到多路召回与 RRF
> **现状说明(2026-09-29)**:本文的排序判断框架(多路召回、RRF、Rerank 选型)仍是理解
> py-rag 服务端检索设计的背景材料;但 RRF 融合、BM25、rerank 的**实现**已下沉 py-rag 服务端,
> Java 侧不再有 `RrfFusion` / `MilvusHybridKnowledgeStore`。
> 当前架构见 [../../architecture/RAG知识检索架构.md](../../architecture/RAG知识检索架构.md)。
**日期**:2026-07-27
**范围**:知识检索排序、多路召回、分数融合、Rerank 选型
**读者**:需要在 Agent 系统里落地 RAG,而不是只做 Demo 问答的工程同学