liveBook · Manning

G-retrieval

As explained in Retrieval Augmented Generation, The Foundational Ideas

G-Retrieval is a graph-based RAG stage that anchors a query to graph nodes and traverses relationships to gather evidence.

G-Retrieval is the online retrieval stage of graph-based RAG. Instead of returning independent top-k chunks through similarity search, it anchors a query to graph nodes and traverses connected edges to gather a coherent evidence subgraph. section 12.1.1

The workflow separates query grounding from evidence gathering: identify nodes named or implied by the query, then walk outward through relationships. This changes the context shape supplied to generation from disconnected passages to connected evidence. section 12.1.1

A typical G-Retrieval path consists of:

Activation spreading in HippoRAG

For “Which Stanford professor works on Alzheimer's?”, HippoRAG extracts Stanford and Alzheimer's as seed nodes, then uses personalized PageRank to spread activation through the knowledge graph. The resulting connected evidence supports multi-hop retrieval rather than simple keyword matching. section 12.3.2

Figure 12.2 The same corpus and the same query, run through both workflows.

Compare standard RAG's independent top-k passages with graph-based RAG's node-grounded traversal and connected subgraph. section 12.1.1

How Retrieval Augmented Generation, The Foundational Ideas covers it

How deep it goes
Explains node anchoring, edge traversal, and PPR activation, but defers detailed traversal mechanics beyond the described workflow.
Examples
Uses a Stanford-and-Alzheimer’s query and contrasts independent passages with a connected evidence subgraph.
What it assumes
Assumes familiarity with standard RAG, while distinguishing its chunk-and-embed retrieval from graph-based grounding and traversal.
What you can do after
Apply it via graph-based RAG by grounding queries on nodes and traversing relationships to retrieve multi-hop evidence.
How it connects
Connects graph-based RAG with G-Indexing, G-Generation, knowledge graphs, similarity search, and personalized PageRank.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Retrieval Augmented Generation, The Foundational Ideas. The section links point to the parts of the book it draws on.