RAGCache: Efficient Knowledge Caching for Retrieval-Augmented Generation

arXiv:2404.12457v1 Announce Type: cross
Abstract: Retrieval-Augmented Generation (RAG) has shown significant improvements in various natural language processing tasks by integrating the strengths of large language models (LLMs) and external knowledge databases. However, RAG introduces long sequence generation and leads to high computation and memory costs. We propose Thoth, a novel multilevel dynamic caching system tailored for RAG. Our analysis benchmarks current RAG systems, pinpointing the performance bottleneck (i.e., long sequence due to knowledge injection) and optimization opportunities (i.e., caching knowledge's …

abstract arxiv caching computation costs cs.cl cs.dc cs.lg databases dynamic however improvements knowledge language language models language processing large language large language models leads llms memory natural natural language natural language processing novel processing rag retrieval retrieval-augmented tasks type