Towards the Anonymization of the Language Modeling

TL;DR

This paper introduces privacy-preserving methods for language model anonymization, using masking and causal modeling techniques to prevent memorization of sensitive data while maintaining utility.

Contribution

It proposes novel MLM and CLM approaches for anonymizing language models, specifically tailored to protect sensitive information in medical datasets.

Findings

Models effectively prevent memorization of personal identifiers.

Proposed methods maintain high utility while enhancing privacy.

Evaluation on medical data shows promising privacy-utility tradeoffs.

Abstract

Rapid advances in Natural Language Processing (NLP) have revolutionized many fields, including healthcare. However, these advances raise significant privacy concerns, especially when pre-trained models fine-tuned and specialized on sensitive data can memorize and then expose and regurgitate personal information. This paper presents a privacy-preserving language modeling approach to address the problem of language models anonymization, and thus promote their sharing. Specifically, we propose both a Masking Language Modeling (MLM) methodology to specialize a BERT-like language model, and a Causal Language Modeling (CLM) methodology to specialize a GPT-like model that avoids the model from memorizing direct and indirect identifying information present in the training data. We have comprehensively evaluated our approaches using a medical dataset and compared them against different baselines. Our results indicate that by avoiding memorizing both direct and indirect identifiers during model specialization, our masking and causal language modeling schemes offer a good tradeoff for maintaining high privacy while retaining high utility.