Adaptive Text Anonymization: Learning Privacy-Utility Trade-offs via Prompt Optimization

2026-02-24

AI Summary

Existing text anonymization methods predominantly rely on static strategies, struggling to dynamically balance privacy preservation with data utility and failing to accommodate the diverse requirements across domains, privacy objectives, and downstream tasks. This work proposes the first framework that formulates anonymization as a learnable, adaptive task, leveraging task-specific prompts to automatically generate context-aware anonymization instructions that guide large language models in flexibly adjusting their anonymization strategies. Built upon open-source language models, the approach is both computationally efficient and highly scalable. Evaluated on a benchmark spanning five distinct domains, the method consistently outperforms existing baselines, achieving performance comparable to that of large closed-source models while significantly improving computational efficiency.

Technology Category

• Machine Learning: Privacy
• Natural Language Processing: (Large) Language Models
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

Anonymizing textual documents is a highly context-sensitive problem: the appropriate balance between privacy protection and utility preservation varies with the data domain, privacy objectives, and downstream application. However, existing anonymization methods rely on static, manually designed strategies that lack the flexibility to adjust to diverse requirements and often fail to generalize across domains. We introduce adaptive text anonymization, a new task formulation in which anonymization strategies are automatically adapted to specific privacy-utility requirements. We propose a framework for task-specific prompt optimization that automatically constructs anonymization instructions for language models, enabling adaptation to different privacy goals, domains, and downstream usage patterns. To evaluate our approach, we present a benchmark spanning five datasets with diverse domains, privacy constraints, and utility objectives. Across all evaluated settings, our framework consistently achieves a better privacy-utility trade-off than existing baselines, while remaining computationally efficient and effective on open-source language models, with performance comparable to larger closed-source models. Additionally, we show that our method can discover novel anonymization strategies that explore different points along the privacy-utility trade-off frontier.

Problem

• text anonymization
• privacy-utility trade-off
• adaptive anonymization
• context-sensitive privacy
• domain generalization

Innovation

• adaptive text anonymization
• prompt optimization
• privacy-utility trade-off
• language models
• automatic anonymization