HybridRAG：整合知识图谱与向量检索增强生成以实现高效信息抽取
HybridRAG: Integrating Knowledge Graphs and Vector Retrieval Augmented Generation for Efficient Information Extraction
AI总结：
本文提出 HybridRAG，融合 GraphRAG 与 VectorRAG，从向量数据库和知识图谱共同检索上下文，提升金融文档问答的信息抽取准确率与生成质量。
摘要：
从金融应用中产生的非结构化文本数据（例如财报电话会议纪要）里抽取并解释复杂信息，即使采用当前最佳实践，即使用检索增强生成（RAG）——指利用向量数据库进行信息检索的 VectorRAG 技术——也会因领域专用术语和文档复杂格式等挑战，而给大型语言模型（LLM）带来巨大困难。我们提出一种名为 HybridRAG 的新颖组合方法，将基于知识图谱（KG）的 RAG 技术（称为 GraphRAG）与 VectorRAG 技术相结合，以增强用于从金融文档中抽取信息的问答（Q&A）系统；结果表明，该方法能够生成准确且符合上下文的答案。我们在一组金融财报电话会议纪要文档上进行实验，这些文档采用 Q&A 格式，因此天然提供了一组真实问答对。实验显示，在检索阶段和生成阶段分别以检索准确率和答案生成质量评估时，同时从向量数据库和 KG 检索上下文的 HybridRAG 优于单独使用的传统 VectorRAG 和 GraphRAG。所提出的技术还具有金融领域之外的应用价值。
英文摘要：
Extraction and interpretation of intricate information from unstructured text data arising in financial applications, such as earnings call transcripts, present substantial challenges to large language models (LLMs) even using the current best practices to use Retrieval Augmented Generation (RAG) (referred to as VectorRAG techniques which utilize vector databases for information retrieval) due to challenges such as domain specific terminology and complex formats of the documents. We introduce a novel approach based on a combination, called HybridRAG, of the Knowledge Graphs (KGs) based RAG techniques (called GraphRAG) and VectorRAG techniques to enhance question-answer (Q&A) systems for information extraction from financial documents that is shown to be capable of generating accurate and contextually relevant answers. Using experiments on a set of financial earning call transcripts documents which come in the form of Q&A format, and hence provide a natural set of pairs of ground-truth Q&As, we show that HybridRAG which retrieves context from both vector database and KG outperforms both traditional VectorRAG and GraphRAG individually when evaluated at both the retrieval and generation stages in terms of retrieval accuracy and answer generation. The proposed technique has applications beyond the financial domain
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。
