Towards Data Science June 24, 2026 By angela shi

Finding the right anchors for RAG: keyword, embedding, and TOC signals in parallel

Read the original on Towards Data Science →

Enterprise Document Intelligence [Vol. 1 #7B] - Retrieval is filtering on structured tables: keywords first, TOC second, embeddings last The post Finding the right anchors for RAG: keyword, embedding, and TOC signals in parallel appeared first on Towards Data Science .

Summary generated by The Flow from the publisher's feed. The full article lives at Towards Data Science.

Towards Data Science

Jun 24

Anchor Detection for RAG: Parallel Detectors, Then One LLM Call at the End

Enterprise Document Intelligence [Vol. 1 #7B] - Retrieval is filtering on structured tables: keywords first, TOC second, embeddings last The post Anchor Detection for RAG: Parallel Detectors, Then One LLM Call at the End appeared first on Towards Data Science .

By angela shi

llms rag

Towards Data Science

Jun 23

Retrieval Is Filtering, Not Search: A Mental Model for Enterprise RAG

Enterprise Document Intelligence [Vol. 1 #7A] - Stop searching strings.

By angela shi

rag

Towards Data Science

Jul 9

Loop Engineering for Hierarchical Retrieval: Reading a Long Document by Its Table of Contents

Enterprise Document Intelligence [Vol. 1 #7quater] - A 492-page document has a 358-entry table of contents.

By Kezhan Shi

Towards Data Science

May 30

Embeddings Aren’t Magic: The Predictable Failure Modes of RAG Retrieval

Enterprise Document Intelligence [Vol. 1 #2] Why the same vector search that handles synonyms and paraphrase silently fails on negation, exact identifiers, and your company’s acronyms, and what to use when it does.

By angela shi

rag

Towards Data Science

Jul 17

A Production RAG Pipeline in Action: Every Answer Typed and Cited

Enterprise Document Intelligence [Vol. 1 #9B] - One call wires the four upgraded bricks together, run on a paper, a NIST standard, and a report with a broken TOC The post A Production RAG Pipeline in Action: Every Answer Typed and Cited appeared first on Towards Data Science .

By angela shi

rag

Towards Data Science

Jun 25

An LLM as arbiter in RAG retrieval: picking the right candidate with reasons

Enterprise Document Intelligence [Vol. 1 #7C] - One LLM call ranks the candidates with reasons.

By angela shi

llms rag

Related stories

Anchor Detection for RAG: Parallel Detectors, Then One LLM Call at the End

Retrieval Is Filtering, Not Search: A Mental Model for Enterprise RAG

Loop Engineering for Hierarchical Retrieval: Reading a Long Document by Its Table of Contents

Embeddings Aren’t Magic: The Predictable Failure Modes of RAG Retrieval

A Production RAG Pipeline in Action: Every Answer Typed and Cited

An LLM as arbiter in RAG retrieval: picking the right candidate with reasons