@inproceedings{hwang-etal-2024-dslr,
title = "{DSLR}: Document Refinement with Sentence-Level Re-ranking and Reconstruction to Enhance Retrieval-Augmented Generation",
author = "Hwang, Taeho and
Jeong, Soyeong and
Cho, Sukmin and
Han, SeungYoon and
Park, Jong",
editor = "Yu, Wenhao and
Shi, Weijia and
Yasunaga, Michihiro and
Jiang, Meng and
Zhu, Chenguang and
Hajishirzi, Hannaneh and
Zettlemoyer, Luke and
Zhang, Zhihan",
booktitle = "Proceedings of the 3rd Workshop on Knowledge Augmented Methods for NLP",
month = aug,
year = "2024",
address = "Bangkok, Thailand",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.knowledgenlp-1.6",
pages = "73--92",
abstract = "Recent advancements in Large Language Models (LLMs) have significantly improved their performance across various Natural Language Processing (NLP) tasks.However, LLMs still struggle with generating non-factual responses due to limitations in their parametric memory.Retrieval-Augmented Generation (RAG) systems address this issue by incorporating external knowledge with a retrieval module.Despite their successes, however, current RAG systems face challenges with retrieval failures and the limited ability of LLMs to filter out irrelevant information.Therefore, in this work, we propose \textit{ \textbf{DSLR}} (\textbf{D}ocument Refinement with \textbf{S}entence-\textbf{L}evel \textbf{R}e-ranking and Reconstruction), an unsupervised framework that decomposes retrieved documents into sentences, filters out irrelevant sentences, and reconstructs them again into coherent passages.We experimentally validate \textit{DSLR} on multiple open-domain QA datasets and the results demonstrate that \textit{DSLR} significantly enhances the RAG performance over conventional fixed-size passage.Furthermore, our \textit{DSLR} enhances performance in specific, yet realistic scenarios without the need for additional training, providing an effective and efficient solution for refining retrieved documents in RAG systems.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hwang-etal-2024-dslr">
<titleInfo>
<title>DSLR: Document Refinement with Sentence-Level Re-ranking and Reconstruction to Enhance Retrieval-Augmented Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Taeho</namePart>
<namePart type="family">Hwang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Soyeong</namePart>
<namePart type="family">Jeong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sukmin</namePart>
<namePart type="family">Cho</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">SeungYoon</namePart>
<namePart type="family">Han</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jong</namePart>
<namePart type="family">Park</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 3rd Workshop on Knowledge Augmented Methods for NLP</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wenhao</namePart>
<namePart type="family">Yu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Weijia</namePart>
<namePart type="family">Shi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Michihiro</namePart>
<namePart type="family">Yasunaga</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Meng</namePart>
<namePart type="family">Jiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chenguang</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hannaneh</namePart>
<namePart type="family">Hajishirzi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Luke</namePart>
<namePart type="family">Zettlemoyer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhihan</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Bangkok, Thailand</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Recent advancements in Large Language Models (LLMs) have significantly improved their performance across various Natural Language Processing (NLP) tasks.However, LLMs still struggle with generating non-factual responses due to limitations in their parametric memory.Retrieval-Augmented Generation (RAG) systems address this issue by incorporating external knowledge with a retrieval module.Despite their successes, however, current RAG systems face challenges with retrieval failures and the limited ability of LLMs to filter out irrelevant information.Therefore, in this work, we propose DSLR (Document Refinement with Sentence-Level Re-ranking and Reconstruction), an unsupervised framework that decomposes retrieved documents into sentences, filters out irrelevant sentences, and reconstructs them again into coherent passages.We experimentally validate DSLR on multiple open-domain QA datasets and the results demonstrate that DSLR significantly enhances the RAG performance over conventional fixed-size passage.Furthermore, our DSLR enhances performance in specific, yet realistic scenarios without the need for additional training, providing an effective and efficient solution for refining retrieved documents in RAG systems.</abstract>
<identifier type="citekey">hwang-etal-2024-dslr</identifier>
<location>
<url>https://aclanthology.org/2024.knowledgenlp-1.6</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>73</start>
<end>92</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T DSLR: Document Refinement with Sentence-Level Re-ranking and Reconstruction to Enhance Retrieval-Augmented Generation
%A Hwang, Taeho
%A Jeong, Soyeong
%A Cho, Sukmin
%A Han, SeungYoon
%A Park, Jong
%Y Yu, Wenhao
%Y Shi, Weijia
%Y Yasunaga, Michihiro
%Y Jiang, Meng
%Y Zhu, Chenguang
%Y Hajishirzi, Hannaneh
%Y Zettlemoyer, Luke
%Y Zhang, Zhihan
%S Proceedings of the 3rd Workshop on Knowledge Augmented Methods for NLP
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand
%F hwang-etal-2024-dslr
%X Recent advancements in Large Language Models (LLMs) have significantly improved their performance across various Natural Language Processing (NLP) tasks.However, LLMs still struggle with generating non-factual responses due to limitations in their parametric memory.Retrieval-Augmented Generation (RAG) systems address this issue by incorporating external knowledge with a retrieval module.Despite their successes, however, current RAG systems face challenges with retrieval failures and the limited ability of LLMs to filter out irrelevant information.Therefore, in this work, we propose DSLR (Document Refinement with Sentence-Level Re-ranking and Reconstruction), an unsupervised framework that decomposes retrieved documents into sentences, filters out irrelevant sentences, and reconstructs them again into coherent passages.We experimentally validate DSLR on multiple open-domain QA datasets and the results demonstrate that DSLR significantly enhances the RAG performance over conventional fixed-size passage.Furthermore, our DSLR enhances performance in specific, yet realistic scenarios without the need for additional training, providing an effective and efficient solution for refining retrieved documents in RAG systems.
%U https://aclanthology.org/2024.knowledgenlp-1.6
%P 73-92
Markdown (Informal)
[DSLR: Document Refinement with Sentence-Level Re-ranking and Reconstruction to Enhance Retrieval-Augmented Generation](https://aclanthology.org/2024.knowledgenlp-1.6) (Hwang et al., KnowledgeNLP-WS 2024)
ACL