KET-RAG: A Cost-Efficient Multi-Granular Indexing Framework for Graph-RAG
Yiqian Huang, Shiqi Zhang, Xiaokui Xiao

TL;DR
KET-RAG introduces a multi-granular indexing framework for Graph-RAG that significantly reduces indexing costs while maintaining or improving retrieval and generation quality in large language model applications.
Contribution
It proposes a novel multi-granular indexing approach combining a knowledge graph skeleton and a lightweight bipartite graph to enhance efficiency and effectiveness.
Findings
Outperforms competitors in indexing cost and retrieval effectiveness.
Achieves comparable or better retrieval quality than Microsoft's Graph-RAG.
Improves generation quality by up to 32.4% while reducing indexing costs by around 20%.
Abstract
Graph-RAG constructs a knowledge graph from text chunks to improve retrieval in Large Language Model (LLM)-based question answering. It is particularly useful in domains such as biomedicine, law, and political science, where retrieval often requires multi-hop reasoning over proprietary documents. Some existing Graph-RAG systems construct KNN graphs based on text chunk relevance, but this coarse-grained approach fails to capture entity relationships within texts, leading to sub-par retrieval and generation quality. To address this, recent solutions leverage LLMs to extract entities and relationships from text chunks, constructing triplet-based knowledge graphs. However, this approach incurs significant indexing costs, especially for large document collections. To ensure a good result accuracy while reducing the indexing cost, we propose KET-RAG, a multi-granular indexing framework.…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAdvanced Graph Neural Networks · Rough Sets and Fuzzy Logic · Graph Theory and Algorithms
MethodsSparse Evolutionary Training
