多语言PII检测混合方法评估研究
An Evaluation Study of Hybrid Methods for Multilingual PII Detection
AI总结：
提出RECAP混合框架，结合正则表达式与LLM，在13种低资源语言中检测PII，支持300多种实体类型，加权F1分数优于微调NER模型82%和零样本LLM 17%。
摘要：
个人可识别信息（PII）的检测对于隐私合规至关重要，但由于语言多样性和标注数据有限，在低资源语言中仍然具有挑战性。我们提出了RECAP，一种混合框架，将确定性正则表达式与上下文感知的大型语言模型（LLMs）相结合，用于在13个低资源语言环境中进行可扩展的PII检测。RECAP的模块化设计支持超过300种实体类型而无需重新训练，采用三阶段细化流程进行消歧和过滤。使用nervaluate进行基准测试，我们的系统在加权F1分数上比微调的NER模型高出82%，比零样本LLMs高出17%。这项工作为合规重点应用中的高效PII检测提供了一种可扩展且适应性强的解决方案。
英文摘要：
The detection of Personally Identifiable Information (PII) is critical for privacy compliance but remains challenging in low-resource languages due to linguistic diversity and limited annotated data. We present RECAP, a hybrid framework that combines deterministic regular expressions with context-aware large language models (LLMs) for scalable PII detection across 13 low-resource locales. RECAP's modular design supports over 300 entity types without retraining, using a three-phase refinement pipeline for disambiguation and filtering. Benchmarked with nervaluate, our system outperforms fine-tuned NER models by 82% and zero-shot LLMs by 17% in weighted F1-score. This work offers a scalable and adaptable solution for efficient PII detection in compliance-focused applications.
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。
