Enregistré dans:
Détails bibliographiques
Auteurs principaux: Guo, Shuyu, Zhang, Shuo, Ren, Zhaochun
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2507.22931
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866915510320365568
author Guo, Shuyu
Zhang, Shuo
Ren, Zhaochun
author_facet Guo, Shuyu
Zhang, Shuo
Ren, Zhaochun
contents Retrieval-augmented generation (RAG) enhances large language models (LLMs) with external knowledge but incurs significant inference costs due to lengthy retrieved contexts. While context compression mitigates this issue, existing methods apply fixed compression rates, over-compressing simple queries or under-compressing complex ones. We propose Adaptive Context Compression for RAG (ACC-RAG), a framework that dynamically adjusts compression rates based on input complexity, optimizing inference efficiency without sacrificing accuracy. ACC-RAG combines a hierarchical compressor (for multi-granular embeddings) with a context selector to retain minimal sufficient information, akin to human skimming. Evaluated on Wikipedia and five QA datasets, ACC-RAG outperforms fixed-rate methods and matches/unlocks over 4 times faster inference versus standard RAG while maintaining or improving accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2507_22931
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Enhancing RAG Efficiency with Adaptive Context Compression
Guo, Shuyu
Zhang, Shuo
Ren, Zhaochun
Computation and Language
Artificial Intelligence
Retrieval-augmented generation (RAG) enhances large language models (LLMs) with external knowledge but incurs significant inference costs due to lengthy retrieved contexts. While context compression mitigates this issue, existing methods apply fixed compression rates, over-compressing simple queries or under-compressing complex ones. We propose Adaptive Context Compression for RAG (ACC-RAG), a framework that dynamically adjusts compression rates based on input complexity, optimizing inference efficiency without sacrificing accuracy. ACC-RAG combines a hierarchical compressor (for multi-granular embeddings) with a context selector to retain minimal sufficient information, akin to human skimming. Evaluated on Wikipedia and five QA datasets, ACC-RAG outperforms fixed-rate methods and matches/unlocks over 4 times faster inference versus standard RAG while maintaining or improving accuracy.
title Enhancing RAG Efficiency with Adaptive Context Compression
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2507.22931