@inproceedings{chen-etal-2025-drag,
title = "{DRAG}: Distilling {RAG} for {SLM}s from {LLM}s to Transfer Knowledge and Mitigate Hallucination via Evidence and Graph-based Distillation",
author = "Chen, Jennifer and
Myrzakhan, Aidar and
Luo, Yaxin and
Khan, Hassaan Muhammad and
Bsharat, Sondos Mahmoud and
Shen, Zhiqiang",
editor = "Che, Wanxiang and
Nabende, Joyce and
Shutova, Ekaterina and
Pilehvar, Mohammad Taher",
booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.acl-long.358/",
doi = "10.18653/v1/2025.acl-long.358",
pages = "7240--7260",
ISBN = "979-8-89176-251-0",
abstract = "Retrieval-Augmented Generation (RAG) methods have proven highly effective for tasks requiring factual consistency and robust knowledge retrieval. However, large-scale RAG systems consume significant computational resources and are prone to generating ``hallucinated'' content from Humans. In this work, we introduce DRAG, a novel framework for distilling RAG knowledge from large-scale Language Models (LLMs) into small LMs (SLMs). Our approach leverages evidence- and knowledge graph{--}based distillation, ensuring that the distilled model retains critical factual knowledge while significantly reducing model size and computational cost. By aligning the smaller model{'}s predictions with a structured knowledge graph and ranked evidence, DRAG effectively mitigates hallucinations and improves factual accuracy. We further present a case demonstrating how our framework mitigates user privacy risks and introduce a corresponding benchmark. Experimental evaluations on multiple benchmarks demonstrate that our method outperforms the prior competitive RAG methods like MiniRAG for SLMs by up to 27.7{\%} using the same models, preserving high-level efficiency and reliability. With DRAG, we provide a practical and resource-efficient roadmap to deploying enhanced retrieval and generation capabilities in small-size LLMs. Code is available at https://github.com/VILA-Lab/DRAG."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chen-etal-2025-drag">
<titleInfo>
<title>DRAG: Distilling RAG for SLMs from LLMs to Transfer Knowledge and Mitigate Hallucination via Evidence and Graph-based Distillation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jennifer</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aidar</namePart>
<namePart type="family">Myrzakhan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yaxin</namePart>
<namePart type="family">Luo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hassaan</namePart>
<namePart type="given">Muhammad</namePart>
<namePart type="family">Khan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sondos</namePart>
<namePart type="given">Mahmoud</namePart>
<namePart type="family">Bsharat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhiqiang</namePart>
<namePart type="family">Shen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wanxiang</namePart>
<namePart type="family">Che</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joyce</namePart>
<namePart type="family">Nabende</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Shutova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="given">Taher</namePart>
<namePart type="family">Pilehvar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Vienna, Austria</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-251-0</identifier>
</relatedItem>
<abstract>Retrieval-Augmented Generation (RAG) methods have proven highly effective for tasks requiring factual consistency and robust knowledge retrieval. However, large-scale RAG systems consume significant computational resources and are prone to generating “hallucinated” content from Humans. In this work, we introduce DRAG, a novel framework for distilling RAG knowledge from large-scale Language Models (LLMs) into small LMs (SLMs). Our approach leverages evidence- and knowledge graph–based distillation, ensuring that the distilled model retains critical factual knowledge while significantly reducing model size and computational cost. By aligning the smaller model’s predictions with a structured knowledge graph and ranked evidence, DRAG effectively mitigates hallucinations and improves factual accuracy. We further present a case demonstrating how our framework mitigates user privacy risks and introduce a corresponding benchmark. Experimental evaluations on multiple benchmarks demonstrate that our method outperforms the prior competitive RAG methods like MiniRAG for SLMs by up to 27.7% using the same models, preserving high-level efficiency and reliability. With DRAG, we provide a practical and resource-efficient roadmap to deploying enhanced retrieval and generation capabilities in small-size LLMs. Code is available at https://github.com/VILA-Lab/DRAG.</abstract>
<identifier type="citekey">chen-etal-2025-drag</identifier>
<identifier type="doi">10.18653/v1/2025.acl-long.358</identifier>
<location>
<url>https://aclanthology.org/2025.acl-long.358/</url>
</location>
<part>
<date>2025-07</date>
<extent unit="page">
<start>7240</start>
<end>7260</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T DRAG: Distilling RAG for SLMs from LLMs to Transfer Knowledge and Mitigate Hallucination via Evidence and Graph-based Distillation
%A Chen, Jennifer
%A Myrzakhan, Aidar
%A Luo, Yaxin
%A Khan, Hassaan Muhammad
%A Bsharat, Sondos Mahmoud
%A Shen, Zhiqiang
%Y Che, Wanxiang
%Y Nabende, Joyce
%Y Shutova, Ekaterina
%Y Pilehvar, Mohammad Taher
%S Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-251-0
%F chen-etal-2025-drag
%X Retrieval-Augmented Generation (RAG) methods have proven highly effective for tasks requiring factual consistency and robust knowledge retrieval. However, large-scale RAG systems consume significant computational resources and are prone to generating “hallucinated” content from Humans. In this work, we introduce DRAG, a novel framework for distilling RAG knowledge from large-scale Language Models (LLMs) into small LMs (SLMs). Our approach leverages evidence- and knowledge graph–based distillation, ensuring that the distilled model retains critical factual knowledge while significantly reducing model size and computational cost. By aligning the smaller model’s predictions with a structured knowledge graph and ranked evidence, DRAG effectively mitigates hallucinations and improves factual accuracy. We further present a case demonstrating how our framework mitigates user privacy risks and introduce a corresponding benchmark. Experimental evaluations on multiple benchmarks demonstrate that our method outperforms the prior competitive RAG methods like MiniRAG for SLMs by up to 27.7% using the same models, preserving high-level efficiency and reliability. With DRAG, we provide a practical and resource-efficient roadmap to deploying enhanced retrieval and generation capabilities in small-size LLMs. Code is available at https://github.com/VILA-Lab/DRAG.
%R 10.18653/v1/2025.acl-long.358
%U https://aclanthology.org/2025.acl-long.358/
%U https://doi.org/10.18653/v1/2025.acl-long.358
%P 7240-7260
Markdown (Informal)
[DRAG: Distilling RAG for SLMs from LLMs to Transfer Knowledge and Mitigate Hallucination via Evidence and Graph-based Distillation](https://aclanthology.org/2025.acl-long.358/) (Chen et al., ACL 2025)
ACL