arXiv 2509.004492025-08-30cs.CL

GOSU：基于全局优化语义单元中心框架的检索增强生成

GOSU: Retrieval-Augmented Generation with Global-Level Optimized Semantic Unit-Centric Framework

School of Future Science and Engineering, Soochow University（未来科学与工程学院，苏州大学）；School of Mathematical Sciences, Soochow University（数学学院，苏州大学）；School of Information Technology (Smart Campus Education Center), Suzhou Institute of Trade & Commerce（信息技术学院（智能校园教育中心），苏州贸易与商业学院）

Xuecheng Zou, Ke Liu, Bingbing Wang, Huafei Deng, Li Zhang, Yu Tang

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对现有基于图的RAG中局部语义单元提取存在歧义、耦合复杂、检索开销高的问题，提出以语义单元为中心的GOSU框架，通过全局合并、分层关键词提取与语义单元补全，提升生成质量。

摘要：

在标准的基于图的检索增强生成（RAG）基础上，异构图和超图的引入旨在通过语义单元（SU）的概念利用多实体间的关系来丰富检索与生成过程。但这也引发了一个关键问题：受限于局部文本块的高层语义单元提取，由于缺乏全局知识或忽略细粒度关系，容易产生歧义、耦合复杂的问题，还会增加检索开销。为解决这些问题，我们提出GOSU——一个以语义单元为中心的RAG框架，可高效执行全局消歧，并利用语义单元捕捉全局上下文下不同节点间的相互关联。在图构建阶段，GOSU对从局部文本块中预提取的语义单元进行全局合并，并指导实体与关系提取，在发现跨文本块的全局语义对象的同时，降低了共指消解的难度。在检索与生成阶段，我们引入了分层关键词提取与语义单元补全机制：前者能发现后者所忽略的细粒度二元关系，后者则可弥补前者缺失的粗粒度n元关系。在多项任务上的评估表明，GOSU在生成质量上优于基线RAG方法。

英文摘要：

Building upon the standard graph-based Retrieval-Augmented Generation (RAG), the introduction of heterogeneous graphs and hypergraphs aims to enrich retrieval and generation by leveraging the relationships between multiple entities through the concept of semantic units (SUs). But this also raises a key issue: The extraction of high-level SUs limited to local text chunks is prone to ambiguity, complex coupling, and increased retrieval overhead due to the lack of global knowledge or the neglect of fine-grained relationships. To address these issues, we propose GOSU, a semantic unit-centric RAG framework that efficiently performs global disambiguation and utilizes SUs to capture interconnections between different nodes across the global context. In the graph construction phase, GOSU performs global merging on the pre-extracted SUs from local text chunks and guides entity and relationship extraction, reducing the difficulty of coreference resolution while uncovering global semantic objects across text chunks. In the retrieval and generation phase, we introduce hierarchical keyword extraction and semantic unit completion. The former uncovers the fine-grained binary relationships overlooked by the latter, while the latter compensates for the coarse-grained n-ary relationships missing from the former. Evaluation across multiple tasks demonstrates that GOSU outperforms the baseline RAG methods in terms of generation quality.

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。