Pré-Publication, Document De Travail Année : 2025

Towards the Anonymization of the Language Modeling

Résumé

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.

Domaines

Fichier principal
Vignette du fichier
2501.02407v2.pdf (2.75 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-05223726 , version 1 (26-08-2025)

Licence

Identifiants

Citer

Juliette Sénéchal, Antoine Boutet, Lucas Magnana, Hélain Zimmermann. Towards the Anonymization of the Language Modeling. 2025. ⟨hal-05223726⟩
184 Consultations
228 Téléchargements

Altmetric

Partager

  • More