大模型
arxiv.org ·
RAG-Safety-Bench：检索增强型大语言模型安全性的可靠评估
该论文提出 RAG-Safety-Bench 基准，用于衡量 RAG 对 LLM 安全性的影响。方法上排除检索器质量的混淆效应，将场景分为四类：非 RAG、含有害请求答案的 oracle 文档、与有害请求相关但无具体答案的文档、随机安全文档，从而隔离不同因素对安全退化的作用。作者在五个开源 LLM 上评测，发现良性能力与不安全能力呈反向关系，基线安全护栏不能保证 RAG 场景下的下游安全，并给出模型层面的证据支持此前“即使良性文档也可能导致检索系统生成不安全内容”的发现。
AI推荐理由 arXiv 新论文（2026-09-10 提交），提出 RAG-Safety-Bench 基准，通过四种条件（非 RAG、含答案的 oracle 文档、相关但无答案文档、随机安全文档）剥离检索器质量这一混淆因素，隔离 RAG 安全退化因素；在五个开源 LLM 上报告结果，给出良性能力与不安全能力呈反向关系、基线安全护栏不保证 RAG 场景下游安全、良性文档也可能导致不安全生成等结论。信息密度高、方法可验证，对 RAG 落地安全评估有直接参考价值；局限是仅覆盖五个开源模型，摘要未给出具体数值与数据集规模。
78
正文
Computer Science > Computation and Language
arXiv:2609.11758 (cs)
[Submitted on 10 Sep 2026]
Title:RAG-Safety-Bench: Reliable Evaluation of Retrieval-Augmented LLM Safety
Authors:Adithiyan Rajan Indira Saravanan, Kathleen C. Fraser
View a PDF of the paper titled RAG-Safety-Bench: Reliable Evaluation of Retrieval-Augmented LLM Safety, by Adithiyan Rajan Indira Saravanan and Kathleen C. Fraser
View PDF
HTML (experimental)
Abstract:Allowing large language models (LLMs) to retrieve information from a set of trusted documents can increase reliability and reduce hallucination. However, recent work has demonstrated that retrieval-augmented generation (RAG) can have unintended side effects on the overall safety of the generated responses, when prompted for harmful or dangerous content. A clearer understanding of the mechanisms leading to this result is needed, as increasing numbers of end users turn to RAG to incorporate corporate documents and knowledge bases into LLM-based systems. We introduce RAG-Safety-Bench, a benchmark to measure the safety impact of RAG on LLM models. By removing the confounding effect of retriever quality, and cleanly separating the problem into four conditions -- non-RAG, RAG with an oracle document containing the answer to the harmful request, RAG with documents related to the harmful request but without the specific answer, and RAG with random, safe documents -- the benchmark isolates the impacts of different factors in the observed safety degradation. We report results across five open-source LLMs, showing an inverse relationship between benign and unsafe capability, strong evidence that baseline safety guardrails do not lead to downstream safety guarantees in the RAG case, and model-specific support for previous findings that even benign documents can lead to unsafe generation in retrieval-enabled systems.
Comments:
Subjects:
Computation and Language (cs.CL); Information Retrieval (cs.IR)
Cite as:
arXiv:2609.11758 [cs.CL]
(or
arXiv:2609.11758v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2609.11758
Focus to learn more
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Kathleen Fraser [view email]
[v1]
Thu, 10 Sep 2026 16:12:25 UTC (2,171 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled RAG-Safety-Bench: Reliable Evaluation of Retrieval-Augmented LLM Safety, by Adithiyan Rajan Indira Saravanan and Kathleen C. Fraser
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.CL
new
|
recent
| 2026-09
Change to browse by:
cs
cs.IR
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
export BibTeX citation
Loading...
BibTeX formatted citation
×
loading...
Data provided by:
Bookmark
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Which authors of this paper are endorsers? |
Disable MathJax (What is MathJax?)