智能体
arxiv.org ·
大型语言模型中的数据安全：风险、防御与方向
这是一篇 LLM 数据安全综述（arXiv:2508.02312，v2 于 2026-09-14 修订，已发表于 J. King Saud Univ. Comput. Inf. Sci.）。文章指出 LLM 依赖海量且未经筛选的训练数据，因而面临有害数据导致的有毒输出与幻觉，以及提示注入、数据投毒等数据驱动攻击。综述回顾了对抗训练、数据清洗、输出护栏、RLHF、数据增强、RAG/Agent 防御等现有防御策略，分类整理了用于评估鲁棒性与安全性的数据集，并提出数据溯源与可追踪、可验证机器遗忘、安全模型更新、标准化评测框架、可解释性驱动的安全分析和治理框架等未来方向。
AI推荐理由 arXiv 综述论文（v2，2026-09-14 修订，已发表于 J. King Saud Univ. Comput. Inf. Sci.），系统梳理 LLM 数据安全风险（提示注入、数据投毒、有害数据导致的有毒输出与幻觉）及防御手段（对抗训练、数据清洗、输出护栏、RLHF、数据增强、RAG/Agent 防御），并整理相关评测数据集，提出数据溯源、可验证机器遗忘、安全模型更新、标准化评测、可解释安全分析与治理框架等方向。信息密度高、来源可验证，但属综述性质，无新实验或新方法，对具体工程落地的直接增量有限。
72
正文
Computer Science > Cryptography and Security
arXiv:2508.02312 (cs)
[Submitted on 4 Aug 2025 (v1), last revised 14 Sep 2026 (this version, v2)]
Title:Data Security in Large Language Models: Risks, Defense, and Directions
Authors:Kang Chen, Xiuze Zhou, Yuanhui Yu, Yuanguo Lin, Hefeng Chen, Congyu Cai, Li Shen
View a PDF of the paper titled Data Security in Large Language Models: Risks, Defense, and Directions, by Kang Chen and 5 other authors
View PDF
HTML (experimental)
Abstract:Large Language Models (LLMs), now a foundation in advancing natural language processing, power applications such as text generation, machine translation, and conversational systems. Despite their transformative potential, these models inherently rely on massive amounts of training data, often collected from diverse and uncurated sources, which exposes them to serious data security risks. Harmful or malicious data can compromise model behavior, leading to toxic outputs or hallucinations, while also creating vulnerabilities to data-driven attacks such as prompt injection and data poisoning. As LLMs continue to be integrated into critical real-world systems, understanding and addressing these data-centric security risks is imperative to safeguard user trust and system reliability. This survey offers a comprehensive overview of the main data security risks facing LLMs and reviews current defense strategies, including adversarial training, data cleaning, output guardrails, Reinforcement Learning from Human Feedback (RLHF), data augmentation, and Retrieval-Augmented Generation (RAG)/agent defenses. Additionally, we categorize and analyze relevant datasets used for assessing robustness and security across different domains, providing guidance for future research. Finally, we highlight key research directions that focus on data provenance and traceability, verifiable machine forgetting, secure model updates, standardized evaluation framework, explainability-driven security analysis, and effective governance frameworks, aiming to promote the safe and responsible development of LLM technology. This work seeks to inform researchers, practitioners, and policymakers, driving progress toward data security in LLMs.
Comments:
Subjects:
Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
Cite as:
arXiv:2508.02312 [cs.CR]
(or
arXiv:2508.02312v2 [cs.CR] for this version)
https://doi.org/10.48550/arXiv.2508.02312
Focus to learn more
arXiv-issued DOI via DataCite
Journal reference:
J. King Saud Univ. Comput. Inf. Sci. 38, 768 (2026)
Related DOI:
https://doi.org/10.1007/s44443-026-01200-9
Focus to learn more
DOI(s) linking to related resources
Submission history
From: Kang Chen [view email]
[v1]
Mon, 4 Aug 2025 11:28:34 UTC (1,764 KB)
[v2]
Mon, 14 Sep 2026 12:03:44 UTC (682 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Data Security in Large Language Models: Risks, Defense, and Directions, by Kang Chen and 5 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.CR
new
|
recent
| 2025-08
Change to browse by:
cs
cs.AI
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?)