不同语言影响大模型隐私泄露程度，意大利语最危险，法语最安全。
The Model's Language Matters: A Comparative Privacy Analysis of LLMs
- 对比英、西、法、意四种语言的医疗模型隐私风险
- 意大利语因冗余度高导致泄漏最强，法语因构词复杂更难被攻破
- 首次量化证明语言结构直接影响隐私安全性，适合多语言系统设计者
大型语言模型在处理多语言敏感数据时面临重大隐私风险。本文针对英语、西班牙语、法语和意大利语的医疗语料训练的LLM进行分析，量化六种语言特征，评估三类攻击：信息提取、反事实记忆和成员推断。结果表明，隐私脆弱性随语言冗余度和分词粒度上升而增强：意大利语泄露最严重，英语成员区分性更高；法语与西班牙语因构词复杂表现出更强韧性。研究首次提供定量证据，证实语言结构显著影响隐私泄露，强调部署中需采用语言感知的隐私保护机制。
原文摘要 · Abstract (English)
Large Language Models (LLMs) are increasingly deployed across multilingual applications that handle sensitive data, yet their scale and linguistic variability introduce major privacy risks. Mostly evaluated for English, this paper investigates how language structure affects privacy leakage in LLMs trained on English, Spanish, French, and Italian medical corpora. We quantify six linguistic indicators and evaluate three attack vectors: extraction, counterfactual memorization, and membership inference. Results show that privacy vulnerability scales with linguistic redundancy and tokenization granularity: Italian exhibits the strongest leakage, while English shows higher membership separability. In contrast, French and Spanish display greater resilience due to higher morphological complexity. Overall, our findings provide the first quantitative evidence that language matters in privacy leakage, underscoring the need for language-aware privacy-preserving mechanisms in LLM deployments.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。