发表机构
ETH Zurich; Columbia University; National University of Singapore（苏黎世联邦理工学院; 哥伦比亚大学; 新加坡国立大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
RAG-Stack是可高效发现RAG质量-性能帕累托前沿的框架，其含三个核心组件，在相同优化迭代下，该框架发现的帕累托前沿覆盖的归一化质量-性能空间比最先进方法多52.5%-153.2%。
AI 中文摘要
检索增强生成（Retrieval-augmented generation，RAG）通过从数据库中检索信息来增强大语言模型（large language model，LLM）的生成，已成为知识密集型应用的广泛使用方法。然而，现代RAG系统存在大量配置选择，如检索索引、模型选择以及模型调用检索的方式，每种配置都会在答案质量与服务性能之间产生不同的权衡，这使得为特定应用部署选择最优设置颇具挑战。本文提出RAG-Stack，这是一个可在各类RAG应用与服务系统中高效发现质量-性能帕累托前沿的框架。RAG-Stack包含三个核心组件：RAG-PE，一种迭代设计空间探索算法，用于选择待评估的下一个RAG配置；RAG-IR，适用于各类RAG算法的工作负载抽象；以及RAG-CM，一种性能模型，可预测给定硬件上的最优部署与服务性能。这些组件协同工作，使RAG-Stack无需部署每个候选即可搜索联合算法-系统配置空间，还能将现有帕累托前沿迁移至新的服务系统。在各类数据集上，当优化迭代次数相同时，相较于在相同RAG设计空间上评估的最先进配置搜索方法，RAG-Stack发现的帕累托前沿覆盖了归一化质量-性能空间的52.5%至153.2%更多区域。
英文摘要
Retrieval-augmented generation (RAG), which augments large language model (LLM) generation with information retrieved from databases, has become a widely used approach for knowledge-intensive applications. Modern RAG systems, however, expose many configuration choices, such as retrieval indexes, model selections, and how models invoke retrieval. Each configuration yields a different trade-off between answer quality and serving performance, making it challenging to choose the optimal setting for a specific application deployment. We present RAG-Stack, a framework for efficiently discovering quality-performance Pareto frontiers across diverse RAG applications and serving systems. RAG-Stack consists of RAG-PE, an iterative design-space exploration algorithm that selects the next RAG configuration to evaluate; RAG-IR, a workload abstraction for diverse RAG algorithms; and RAG-CM, a performance model that predicts the optimal deployment and serving performance on the given hardware. Together, these components allow RAG-Stack to search the joint algorithm-system configuration space without deploying every candidate and to transfer an existing Pareto frontier to a new serving system. Given the same number of optimization iterations across diverse datasets, the Pareto frontiers found by RAG-Stack cover 52.5% to 153.2% more of the normalized quality-performance space than those found by state-of-the-art configuration-search methods evaluated over the same RAG design space.
