arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.08545cs.IR

DA-RAG:动态属性社区搜索用于检索增强生成

DA-RAG: Dynamic Attributed Community Search for Retrieval-Augmented Generation

Xingyuan Zeng, Zuohan Wu, Yue Wang, Chen Zhang, Quanming Yao, Libin Zheng, Jian Yin

AI总结:

DA-RAG通过动态属性社区搜索提升检索增强生成的效果,有效处理动态复杂查询,减少计算与经济成本。

摘要:

鉴于其前所未有的理解能力,大语言模型(LLMs)已成为现代网络搜索引擎不可或缺的组成部分。从技术角度来看,这种整合代表了检索增强生成(RAG),通过将LLMs与外部知识库结合来增强其能力。在这一背景下,图基于RAG(G-RAG)是一种常见的技术方法。然而,当前的G-RAG方法经常低估了图拓扑,主要关注低阶结构或预计算的静态社区。这种限制影响了它们在处理动态和复杂查询时的有效性。因此,我们提出了DA-RAG,利用属性社区搜索(ACS)根据查询问题动态提取相关子图。DA-RAG捕捉高阶图结构,允许检索自补足的知识。此外,DA-RAG配备了面向块层的图索引,这使得多粒度检索更加高效,同时显著降低了计算和经济成本。我们在多个数据集上评估了DA-RAG,证明其在四个指标上的头对头比较中,比现有RAG方法高出多达40%,同时将索引构建时间和令牌开销分别减少多达37%和41%。

英文摘要:

Owing to their unprecedented comprehension capabilities, large language models (LLMs) have become indispensable components of modern web search engines. From a technical perspective, this integration represents retrieval-augmented generation (RAG), which enhances LLMs by grounding them in external knowledge bases. A prevalent technical approach in this context is graph-based RAG (G-RAG). However, current G-RAG methodologies frequently underutilize graph topology, predominantly focusing on low-order structures or pre-computed static communities. This limitation affects their effectiveness in addressing dynamic and complex queries. Thus, we propose DA-RAG, which leverages attributed community search (ACS) to extract relevant subgraphs based on the queried question dynamically. DA-RAG captures high-order graph structures, allowing for the retrieval of self-complementary knowledge. Furthermore, DA-RAG is equipped with a chunk-layer oriented graph index, which facilitates efficient multi-granularity retrieval while significantly reducing both computational and economic costs. We evaluate DA-RAG on multiple datasets, demonstrating that it outperforms existing RAG methods by up to 40% in head-to-head comparisons across four metrics while reducing index construction time and token overhead by up to 37% and 41%, respectively.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑