Searching for Best Practices in Retrieval-Augmented Generation

TL;DR

This paper investigates various retrieval-augmented generation strategies to optimize performance and efficiency, and explores multimodal retrieval techniques to improve visual question answering and content generation.

Contribution

It systematically analyzes existing RAG approaches, proposes optimal deployment strategies, and demonstrates the benefits of multimodal retrieval for enhanced multimodal content generation.

Findings

Optimal RAG strategies balance performance and efficiency.

Multimodal retrieval significantly improves visual question-answering.

Retrieval as generation accelerates multimodal content creation.

Abstract

Retrieval-augmented generation (RAG) techniques have proven to be effective in integrating up-to-date information, mitigating hallucinations, and enhancing response quality, particularly in specialized domains. While many RAG approaches have been proposed to enhance large language models through query-dependent retrievals, these approaches still suffer from their complex implementation and prolonged response times. Typically, a RAG workflow involves multiple processing steps, each of which can be executed in various ways. Here, we investigate existing RAG approaches and their potential combinations to identify optimal RAG practices. Through extensive experiments, we suggest several strategies for deploying RAG that balance both performance and efficiency. Moreover, we demonstrate that multimodal retrieval techniques can significantly enhance question-answering capabilities about visual inputs and accelerate the generation of multimodal content using a "retrieval as generation" strategy.