Graph-Guided Concept Selection for Efficient Retrieval-Augmented Generation

TL;DR

This paper introduces G2ConS, a novel method that reduces retrieval-augmented generation costs by selecting important concepts and chunks using a graph-guided approach, improving efficiency and answer quality.

Contribution

G2ConS is the first approach to combine concept selection with an LLM-independent graph to enhance retrieval efficiency in RAG systems.

Findings

G2ConS reduces knowledge graph construction costs significantly.

G2ConS improves retrieval effectiveness over baselines.

G2ConS enhances answer quality in real-world datasets.

Abstract

Graph-based RAG constructs a knowledge graph (KG) from text chunks to enhance retrieval in Large Language Model (LLM)-based question answering. It is especially beneficial in domains such as biomedicine, law, and political science, where effective retrieval often involves multi-hop reasoning over proprietary documents. However, these methods demand numerous LLM calls to extract entities and relations from text chunks, incurring prohibitive costs at scale. Through a carefully designed ablation study, we observe that certain words (termed concepts) and their associated documents are more important. Based on this insight, we propose Graph-Guided Concept Selection (G2ConS). Its core comprises a chunk selection method and an LLM-independent concept graph. The former selects salient document chunks to reduce KG construction costs; the latter closes knowledge gaps introduced by chunk selection at zero cost. Evaluations on multiple real-world datasets show that G2ConS outperforms all baselines in construction cost, retrieval effectiveness, and answering quality.