arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2603.17217cs.CLcs.AIcs.LG
收藏

匿名由设计:一个基于LLM的隐私保护文本框架

Anonymous-by-Construction: An LLM-Driven Framework for Privacy-Preserving Text

Veritran;University of Buenos Aires(布宜诺斯艾利斯大学);University of San Andrés(圣安德烈斯大学)

Federico Albanese, Pablo Ronco, Nicolás D'Ippolito

AI总结:

本文提出基于LLM的匿名化框架,通过本地LLM实现文本隐私保护,同时保持语义和任务相关性,优于现有方法。

摘要:

人工智能的负责任使用要求我们在不损害数据有用性的情况下保护敏感信息,这在大语言模型时代变得尤为紧迫。我们通过一个本地化、基于LLM的替换流程来解决这一挑战,该流程通过将个人可识别信息(PII)替换为真实且类型一致的替代项来匿名化文本。该方法完全在组织边界内执行,利用本地LLM,防止数据流出的同时保持流畅性和任务相关语义。我们对Action-Based Conversation Dataset进行了系统性、多指标、跨技术评估,与行业标准(Microsoft Presidio和Google DLP)和最先进的方法(ZSTS,红制和红制加替换变体)进行基准测试。我们的协议通过在净化文本上微调紧凑编码器(BERT+LoRA)获得了一种生命周期准备的指标,共同测量隐私、语义效用和隐私下的训练性。此外,我们通过在回答LLM之前插入本地匿名化层来评估代理问答性能,并评估其响应质量。这种中间、类型保持的替换阶段确保敏感内容不会暴露给第三方API,从而在不损害保密性的情况下实现问答代理的负责任部署。我们的方法在隐私、最小主题漂移、强事实效用和低训练损失方面均达到最先进水平,优于基于规则的方法和命名实体识别(NER)基线以及ZSTS变体在隐私-效用-训练性前沿上。这些结果表明,本地LLM替换生成的匿名语料库既负责任又在操作上有价值:安全用于代理流程,适合下游微调且降解有限。

英文摘要:

Responsible use of AI demands that we protect sensitive information without undermining the usefulness of data, an imperative that has become acute in the age of large language models. We address this challenge with an on-premise, LLM-driven substitution pipeline that anonymizes text by replacing personally identifiable information (PII) with realistic, type-consistent surrogates. Executed entirely within organizational boundaries using local LLMs, the approach prevents data egress while preserving fluency and task-relevant semantics. We conduct a systematic, multi-metric, cross-technique evaluation on the Action-Based Conversation Dataset, benchmarking against industry standards (Microsoft Presidio and Google DLP) and a state-of-the-art approach (ZSTS, in redaction-only and redaction-plus-substitution variants). Our protocol jointly measures privacy, semantic utility, and trainability under privacy via a lifecycle-ready criterion obtained by fine-tuning a compact encoder (BERT+LoRA) on sanitized text. In addition, we assess agentic Q&A performance by inserting an on-premise anonymization layer before the answering LLM and evaluating the quality of its responses. This intermediate, type-preserving substitution stage ensures that no sensitive content is exposed to third-party APIs, enabling responsible deployment of Q\&A agents without compromising confidentiality. Our method attains state-of-the-art privacy, minimal topical drift, strong factual utility, and low trainability loss, outperforming rule-based approaches and named-entity recognition (NER) baselines and ZSTS variants on the combined privacy--utility--trainability frontier. These results show that local LLM substitution yields anonymized corpora that are both responsible to use and operationally valuable: safe for agentic pipelines and suitable for downstream fine-tuning with limited degradation.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑