PISCO: Pretty Simple Compression for Retrieval-Augmented Generation

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Louis, Maxime, Déjean, Hervé, Clinchant, Stéphane
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909466854686720
author Louis, Maxime
Déjean, Hervé
Clinchant, Stéphane
author_facet Louis, Maxime
Déjean, Hervé
Clinchant, Stéphane
contents Retrieval-Augmented Generation (RAG) pipelines enhance Large Language Models (LLMs) by retrieving relevant documents, but they face scalability issues due to high inference costs and limited context size. Document compression is a practical solution, but current soft compression methods suffer from accuracy losses and require extensive pretraining. In this paper, we introduce PISCO, a novel method that achieves a 16x compression rate with minimal accuracy loss (0-3%) across diverse RAG-based question-answering (QA) tasks. Unlike existing approaches, PISCO requires no pretraining or annotated data, relying solely on sequence-level knowledge distillation from document-based questions. With the ability to fine-tune a 7-10B LLM in 48 hours on a single A100 GPU, PISCO offers a highly efficient and scalable solution. We present comprehensive experiments showing that PISCO outperforms existing compression models by 8% in accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2501_16075
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PISCO: Pretty Simple Compression for Retrieval-Augmented Generation
Louis, Maxime
Déjean, Hervé
Clinchant, Stéphane
Computation and Language
Artificial Intelligence
Information Retrieval
Retrieval-Augmented Generation (RAG) pipelines enhance Large Language Models (LLMs) by retrieving relevant documents, but they face scalability issues due to high inference costs and limited context size. Document compression is a practical solution, but current soft compression methods suffer from accuracy losses and require extensive pretraining. In this paper, we introduce PISCO, a novel method that achieves a 16x compression rate with minimal accuracy loss (0-3%) across diverse RAG-based question-answering (QA) tasks. Unlike existing approaches, PISCO requires no pretraining or annotated data, relying solely on sequence-level knowledge distillation from document-based questions. With the ability to fine-tune a 7-10B LLM in 48 hours on a single A100 GPU, PISCO offers a highly efficient and scalable solution. We present comprehensive experiments showing that PISCO outperforms existing compression models by 8% in accuracy.
title PISCO: Pretty Simple Compression for Retrieval-Augmented Generation
topic Computation and Language
Artificial Intelligence
Information Retrieval
url https://arxiv.org/abs/2501.16075