Author

aiterms

What is graph RAG?

Graph RAG, short for Graph Retrieval-Augmented Generation, is a way to help an LLM answer questions by retrieving information from a graph of connected entities and relationships instead of, or in addition to, plain text chunks.

Why it matters

Classic RAG works well when the answer lives in a few relevant text passages. But some questions are really about relationships: who depends on whom, how events connect, which product is linked to which system, or how a concept appears across many documents.

Graph RAG is useful when you need:

• multi-hop reasoning across connected facts
• better navigation of messy enterprise knowledge
• answers that depend on entities, links, and context, not just keyword similarity

In practice, teams reach for graph RAG when simple vector search starts missing important context or returning isolated snippets that don’t show the full picture.

How it works

At a high level, graph RAG adds a graph layer to retrieval.

1. Build a graph from your knowledge

   • Nodes might be people, products, documents, tickets, tables, or concepts.
   • Edges represent relationships such as “mentions,” “depends on,” “owned by,” or “causes.”

2. Retrieve through the graph

   • Given a question, the system finds relevant starting nodes.
   • It then expands to nearby connected nodes, often following a few hops to collect related context.

3. Feed the gathered context to the LLM

   • The model gets a curated set of nodes, edges, and supporting text.
   • It uses that structured context to generate an answer, sometimes with citations back to source material.

A common pattern is to combine graph search with text retrieval: the graph helps you find the right neighborhood, and the source documents provide the evidence.

Tiny concrete example

Question: “Which services are affected if payment gateway A goes down?”

A graph might contain:

• payment gateway A -> used by -> checkout service
• checkout service -> calls -> orders service
• orders service -> depends on -> billing service

The retriever can walk those links and provide the LLM with the affected services and supporting documents, instead of only a handful of isolated text chunks mentioning “payment gateway A.”

Common pitfalls / when NOT to use it

• Overkill for simple FAQ search. If users mostly ask standalone questions with answers in one or two passages, plain RAG is often enough.
• Garbage-in, garbage-out. If the graph is incomplete or noisy, the retrieved context will be incomplete or noisy too.
• Graph construction is the hard part. Extracting reliable entities and relationships from documents is often more expensive than people expect.
• Not automatically better. Graph RAG can improve recall for connected questions, but it adds complexity in indexing, retrieval, and maintenance.
• Ambiguous term. “Graph RAG” is not one single standardized algorithm; it’s a family of approaches that use graph-structured retrieval for RAG.

Related terms

• What is retrieval-augmented generation (RAG)?
• What is a vector database?
• What is an embedding?
• What is semantic search?
• What is chunking (and semantic chunking)?
• What is hybrid search (BM25 + vector)?
• What is a reranker / cross-encoder?
• What is top-k retrieval?
• What is a knowledge base for RAG?
• What is contextual / late chunking?
• What is RAG vs fine-tuning?
• What is a vector index (HNSW / IVF)?
• What is cosine similarity?
• What is metadata filtering in retrieval?
• What is query rewriting / expansion?
• What is the lost-in-the-middle problem?
• What is RAG evaluation (faithfulness and relevance)?
• What is a document loader / parser?

同じ著者の記事

What is the lost-in-the-middle problem?

The lost-in-the-middle problem is the tendency of a language model to miss information that appears in the middle of a long context, even when that information is relevant to the answer. This matters any time you stuff a lot of text into an LLM prompt: long documents, chat histories, codebases, retrieval-augmented generation (RAG), or multi-step agent traces. In practice, it means: the model may answer well using the beginning or end of the context, but fail on facts buried in the middle, so

papoo.work

What is query rewriting / expansion?

Query rewriting or query expansion is the process of changing a user’s search query into a better one—usually by adding, removing, or rephrasing terms so a search or retrieval system can find more relevant results. People often search with short, vague, or underspecified queries. Systems that use the raw query alone may miss relevant documents because the wording does not match the target content. Query rewriting/expansion helps when you want better recall, better intent matching, or more robust

papoo.work

What is metadata filtering in retrieval?

Metadata filtering in retrieval is the practice of narrowing a search or vector search to documents that match structured fields like date, author, source, tenant, language, or content type before or during ranking. It solves a very practical problem: not every relevant item is relevant for this user, this time, or this task. If you are building search or retrieval-augmented generation (RAG), metadata filters help you: exclude clearly irrelevant documents early, enforce access control or tenant

papoo.work

What is cosine similarity?

Cosine similarity is a way to measure how similar two vectors are by comparing the angle between them, not their size. It’s useful when you care more about direction than magnitude. In AI and information retrieval, that usually means comparing embeddings, text vectors, or feature vectors to find items that “mean the same thing” even if they have different lengths. You’d reach for cosine similarity when: ranking search results by semantic closeness, comparing document, sentence, or image embe

papoo.work

What is a vector index (HNSW / IVF)?

A vector index is a data structure that helps a system quickly find the most similar embeddings or vectors to a query vector; it is also commonly called an ANN index, short for approximate nearest neighbor index. If you store thousands, millions, or billions of vectors and want “find the closest ones,” a brute-force scan is usually too slow. A vector index lets you trade a little exactness for much faster retrieval, which is why it shows up in search, retrieval-augmented generation (RAG), recomm

papoo.work

What is contextual / late chunking?

Contextual chunking, also called late chunking, is a way to split a document into retrieval chunks after a model has already encoded the whole document, so each chunk keeps some awareness of the surrounding context. Normal chunking cuts text first and then embeds each piece alone. That is simple, but it can make chunks semantically thin: a paragraph may be ambiguous without the surrounding sections, headings, or earlier definitions. Late chunking helps when you want better retrieval quality

papoo.work

What is a knowledge base for RAG?

A knowledge base for RAG is the collection of documents or records a system retrieves from to ground an LLM’s answer in relevant source material. RAG, short for retrieval-augmented generation, works best when the model can search a useful, trustworthy set of content before it writes. That content is the knowledge base. You’d reach for one when the model needs to answer questions from: company docs support articles policies and handbooks product specs research notes tickets or tickets-like record

papoo.work

What is top-k retrieval?

Top-k retrieval is a way to search a collection and return the k most relevant items for a query, instead of trying to sort or inspect everything. Top-k retrieval is the standard first step in search, recommendation, and retrieval-augmented generation (RAG). You use it when you want a fast shortlist: the most relevant documents, passages, products, images, or candidates, without paying the cost of full ranking over the entire corpus. In practice, teams reach for top-k retrieval when: the dat

papoo.work

What is a reranker / cross-encoder?

A reranker, often implemented as a cross-encoder, is a model that takes a query and a candidate result together and scores how well they match. A reranker solves the “good enough first pass, better final answer” problem. In search and retrieval systems, you usually start with a fast retriever that finds a few dozen or few hundred candidate documents. That first stage is optimized for speed, not perfect ranking. A reranker then looks at those candidates more carefully and reorders them so the mos

papoo.work

What is hybrid search (BM25 + vector)?

Hybrid search is a retrieval method that combines keyword search with vector similarity search so you can find documents that match both the exact words a user typed and the broader meaning of the query. Pure keyword search is good when the exact term matters. Pure vector search is good when wording varies but meaning is similar. In practice, many real queries need both. You reach for hybrid search when: users may search with exact product names, error codes, or technical terms the same

papoo.work