ASIRF：一种面向上下文相关敏感信息脱敏的智能体框架
ASIRF: An Agentic Framework for Context-Dependent Sensitive Information Redaction
浏览论文内容
中文总结 AI 辅助
ASIRF是一种智能体框架，通过推理时检索领域特定定义实现无需重训练的敏感信息脱敏，在多数模型-领域组合中召回率优于基线OPF。
中文摘要 AI 辅助
敏感信息由领域和意图定义，而非通用类别，然而诸如隐私过滤器和命名实体识别器等脱敏系统在训练时固定了分类体系，导致每个新领域都需要重新训练。我们提出ASIRF（智能体敏感信息脱敏框架），该框架在推理时根据输入领域从灵活的知识库中检索领域特定的定义，无需重新训练即可适应。我们在十个小型开放权重模型和八个数据集（包括分布外的虚构领域）上评估了两种架构——一种三次调用的多智能体流水线和一种单智能体变体，并以OpenAI隐私过滤器（OPF）作为训练分类器基线进行比较。仅凭每个领域几十条专家编写的定义且无训练数据，ASIRF在80个模型-领域组合中的68个（85%）中，至少通过两种架构之一，其召回率超过OPF，而不足之处主要集中在OPF训练分布所覆盖的领域。
英文摘要
Sensitive information is defined by domain and intent, not a universal category, yet redaction systems such as privacy filters and named-entity recognizers fix a taxonomy at training time, requiring retraining for each new domain. We introduce ASIRF (Agentic Sensitive Information Redaction Framework), which retrieves domain-specific definitions based on the input's domain from a flexible knowledge base at inference time, needing no retraining to adapt. Two architectures, a three-call multi-agent pipeline and a single-agent variant, are evaluated across ten small open-weight models and eight datasets, including out-of-distribution fictional domains, against the OpenAI Privacy Filter (OPF) as a trained-classifier baseline. With only a few dozen expert-authored definitions per domain and no training data, ASIRF's recall exceeds OPF's in 68 of 80 model-domain combinations (85 percent), by at least one of the two architectures, with shortfalls confined mostly to OPF's training-distribution domains.
发表机构
- University of Liverpool（利物浦大学）
- Keele University（基尔大学）
机构由 AI 辅助整理，请以论文原文为准。
