liveBook · Manning

Passage retrieval

As explained in Effective Conversational AI

Passage retrieval is the process of finding text passages relevant to a query from a document collection.

When a chatbot cannot answer reliably, passage retrieval finds relevant text in a document collection and supplies evidence for response generation. In a retrieval-augmented workflow, the system retrieves related passages, generates a candidate answer, checks its support, and may transfer the user to a human agent. section 6.2.2

Checking an uncertain health question

For a question about achalasia, dysphagia, and vaccine side effects, retrieval can return passages defining achalasia, describing vaccine side effects, and connecting dysphagia with esophageal conditions. If the generated answer adds an unsupported claim, the system can avoid presenting it as reliable and connect the user with a specialist. section 6.2.2

How Effective Conversational AI covers it

How deep it goes
Explains retrieval, answer generation, comparison, and transfer decisions, without detailing retrieval algorithms or indexing internals.
Examples
Uses a health question about achalasia, dysphagia, vaccine side effects, retrieved passages, and a specialist-transfer response.
What you can do after
Diagnose unsupported chatbot answers by comparing generated claims with retrieved passages and escalating poorly grounded cases.
How it connects
Connects retrieval-augmented generation with large language models, grounding checks, and human-agent escalation.
Cover of Effective Conversational AI
liveBookmanning.com
Read Effective Conversational AI on liveBook Get the book at manning.com

This explainer was generated by Manning's knowledge graph from the text of Effective Conversational AI. The section links point to the parts of the book it draws on.