Saved in:
Bibliographic Details
Main Authors: Liu, Ziyu, Liu, Yijing, Yuan, Jianfei, Yan, Minzhi, Yue, Le, Xiong, Honghui, Yang, Yi
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.24120
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914117920489472
author Liu, Ziyu
Liu, Yijing
Yuan, Jianfei
Yan, Minzhi
Yue, Le
Xiong, Honghui
Yang, Yi
author_facet Liu, Ziyu
Liu, Yijing
Yuan, Jianfei
Yan, Minzhi
Yue, Le
Xiong, Honghui
Yang, Yi
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2510_24120
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Graph-Guided Concept Selection for Efficient Retrieval-Augmented Generation
Liu, Ziyu
Liu, Yijing
Yuan, Jianfei
Yan, Minzhi
Yue, Le
Xiong, Honghui
Yang, Yi
Machine Learning
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.
title Graph-Guided Concept Selection for Efficient Retrieval-Augmented Generation
topic Machine Learning
url https://arxiv.org/abs/2510.24120