A Survey Of Vector-Indexed And Graph-Based Approaches To Retrieval-Augmented Generation (RAG)

Abstract

: This survey explores the various vector-indexed and graph-based approaches employed in Retrieval-Augmented Generation (RAG) systems, a paradigm that enhances large language models (LLMs) by incorporating external knowledge through retrieval mechanisms. RAG systems enable LLMs to generate more accurate, contextually aware, and informative responses by integrating external documents or structured data during the generation process. We delve into two main techniques: vector-indexed retrieval, which leverages high-dimensional vector embeddings and methods like Hierarchical Navigable Small World Graphs (HNSW) for efficient approximate nearest neighbor search, and graph-based retrieval, which utilizes knowledge graphs to store and query relational data. We examine the key stages involved in these approaches, including data preprocessing, embedding generation, search algorithms, and integration with LLMs. Furthermore, we address the challenges of handling large-scale data and unstructured information. The paper provides a comprehensive overview of the current methods, their strengths, and the limitations of RAG.