@inproceedings{minh-etal-2026-retrieved,
title = "Retrieved But Not Reliable: A Survey on Attacks, and Defenses in Retrieval-Augmented Generation",
author = "Minh, Tran Ta Quang and
Dang, Cuong and
Nguyen, Tuc and
Tran, Khanh-Tung and
Huynh Nguyen, Minh and
Trinh, Chau Nguyen To and
Le, Kien and
Long, Do Xuan and
Zhang, Jiahao and
Wang, Fali and
Nguyen, Hoang D. and
Le, Thanh and
Wang, Suhang",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.1269/",
pages = "26832--26855",
ISBN = "979-8-89176-462-0",
abstract = "Retrieval-Augmented Generation (RAG) enhances large language models by grounding outputs in external knowledge, improving factuality and reducing hallucinations. At the same time, the retrieval-augmented pipeline introduces new robustness and security risks, including corpus poisoning, backdoor attacks, privacy leakage, and fairness violations. Despite rapid progress in this area, existing surveys remain limited in their treatment of attacker objectives, threat models, and stage-specific defenses across the full RAG pipeline. This survey presents a unified and pipeline-aware overview of RAG robustness. We formalize threat models over the corpus, retriever, and generator, and organize attacks into three main objectives: accuracy, privacy, and fairness. We further review defenses from a pipeline-aware perspective, covering the retrieval, rerank, generation, and traceback stages. In addition, we summarize robustness benchmarks and explainability methods for more deeply evaluating and explaining RAG robustness. Finally, we highlight open challenges and future directions for building trustworthy RAG systems. To continuously track developments in this rapidly evolving field, we will also actively update a public repository at https://github.com/coutMinh/A-Survey-on-RAG-Robustness."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="minh-etal-2026-retrieved">
<titleInfo>
<title>Retrieved But Not Reliable: A Survey on Attacks, and Defenses in Retrieval-Augmented Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Tran</namePart>
<namePart type="given">Ta</namePart>
<namePart type="given">Quang</namePart>
<namePart type="family">Minh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cuong</namePart>
<namePart type="family">Dang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tuc</namePart>
<namePart type="family">Nguyen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Khanh-Tung</namePart>
<namePart type="family">Tran</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Minh</namePart>
<namePart type="family">Huynh Nguyen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chau</namePart>
<namePart type="given">Nguyen</namePart>
<namePart type="given">To</namePart>
<namePart type="family">Trinh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kien</namePart>
<namePart type="family">Le</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Do</namePart>
<namePart type="given">Xuan</namePart>
<namePart type="family">Long</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiahao</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fali</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hoang</namePart>
<namePart type="given">D</namePart>
<namePart type="family">Nguyen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Thanh</namePart>
<namePart type="family">Le</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Suhang</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>Retrieval-Augmented Generation (RAG) enhances large language models by grounding outputs in external knowledge, improving factuality and reducing hallucinations. At the same time, the retrieval-augmented pipeline introduces new robustness and security risks, including corpus poisoning, backdoor attacks, privacy leakage, and fairness violations. Despite rapid progress in this area, existing surveys remain limited in their treatment of attacker objectives, threat models, and stage-specific defenses across the full RAG pipeline. This survey presents a unified and pipeline-aware overview of RAG robustness. We formalize threat models over the corpus, retriever, and generator, and organize attacks into three main objectives: accuracy, privacy, and fairness. We further review defenses from a pipeline-aware perspective, covering the retrieval, rerank, generation, and traceback stages. In addition, we summarize robustness benchmarks and explainability methods for more deeply evaluating and explaining RAG robustness. Finally, we highlight open challenges and future directions for building trustworthy RAG systems. To continuously track developments in this rapidly evolving field, we will also actively update a public repository at https://github.com/coutMinh/A-Survey-on-RAG-Robustness.</abstract>
<identifier type="citekey">minh-etal-2026-retrieved</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.1269/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>26832</start>
<end>26855</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Retrieved But Not Reliable: A Survey on Attacks, and Defenses in Retrieval-Augmented Generation
%A Minh, Tran Ta Quang
%A Dang, Cuong
%A Nguyen, Tuc
%A Tran, Khanh-Tung
%A Huynh Nguyen, Minh
%A Trinh, Chau Nguyen To
%A Le, Kien
%A Long, Do Xuan
%A Zhang, Jiahao
%A Wang, Fali
%A Nguyen, Hoang D.
%A Le, Thanh
%A Wang, Suhang
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F minh-etal-2026-retrieved
%X Retrieval-Augmented Generation (RAG) enhances large language models by grounding outputs in external knowledge, improving factuality and reducing hallucinations. At the same time, the retrieval-augmented pipeline introduces new robustness and security risks, including corpus poisoning, backdoor attacks, privacy leakage, and fairness violations. Despite rapid progress in this area, existing surveys remain limited in their treatment of attacker objectives, threat models, and stage-specific defenses across the full RAG pipeline. This survey presents a unified and pipeline-aware overview of RAG robustness. We formalize threat models over the corpus, retriever, and generator, and organize attacks into three main objectives: accuracy, privacy, and fairness. We further review defenses from a pipeline-aware perspective, covering the retrieval, rerank, generation, and traceback stages. In addition, we summarize robustness benchmarks and explainability methods for more deeply evaluating and explaining RAG robustness. Finally, we highlight open challenges and future directions for building trustworthy RAG systems. To continuously track developments in this rapidly evolving field, we will also actively update a public repository at https://github.com/coutMinh/A-Survey-on-RAG-Robustness.
%U https://aclanthology.org/2026.findings-emnlp.1269/
%P 26832-26855
Markdown (Informal)
[Retrieved But Not Reliable: A Survey on Attacks, and Defenses in Retrieval-Augmented Generation](https://aclanthology.org/2026.findings-emnlp.1269/) (Minh et al., Findings 2026)
ACL
- Tran Ta Quang Minh, Cuong Dang, Tuc Nguyen, Khanh-Tung Tran, Minh Huynh Nguyen, Chau Nguyen To Trinh, Kien Le, Do Xuan Long, Jiahao Zhang, Fali Wang, Hoang D. Nguyen, Thanh Le, and Suhang Wang. 2026. Retrieved But Not Reliable: A Survey on Attacks, and Defenses in Retrieval-Augmented Generation. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 26832–26855, Budapest, Hungary. Association for Computational Linguistics.