mlsanityData to 7 Oct 2026

Unsupervised Text Deidentification

arXiv 2210.11528 ↗First posted 20 Oct 2022cs.CLcited by 11

Not on the map. Why?

Abstract (arXiv)

Deidentification seeks to anonymize textual data prior to distribution. Automatic deidentification primarily uses supervised named entity recognition from human-labeled data points. We propose an unsupervised deidentification method that masks words that leak personally-identifying information. The approach utilizes a specially trained reidentification model to identify individuals from redacted personal documents. Motivated by K-anonymity based privacy, we generate redactions that ensure a minimum reidentification rank for the correct profile of the document. To evaluate this approach, we consider the task of deidentifying Wikipedia Biographies, and evaluate using an adversarial reidentification metric. Compared to a set of unsupervised baselines, our approach deidentifies documents more completely while removing fewer words. Qualitatively, we see that the approach eliminates many identifying aspects that would fall outside of the common named entity based approach.

Key facts. Cited by 11 papers in the map (data through 7 Oct 2026). Builds on 15 of its 45 references that are in the map. First posted on arXiv 20 Oct 2022.

Builds on

15 of 45 references.

References · 15

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
  3. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,779
  4. 2019When Does Label Smoothing Help?cited by 985
  5. 2020TAPAS: Weakly Supervised Table Parsing via Pre-trainingcited by 432
  6. 2021Large Language Models Can Be Strong Differentially Private Learnerscited by 405
  7. 2016Understanding Neural Networks through Representation Erasurecited by 223
  8. 2019DeepPrivacy: A Generative Adversarial Network for Face Anonymizationcited by 108
  9. 2022The Text Anonymization Benchmark (TAB): A Dedicated Corpus and Evaluation Framework for Text Anonymizationcited by 94
  10. 2020CIAGAN: Conditional Identity Anonymization Generative Adversarial Networkscited by 88
  11. 2020Learning Variational Word Masks to Improve the Interpretability of Neural Text Classifierscited by 35
  12. 2020Probabilistically Masked Language Model Capable of Autoregressive Generation in Arbitrary Word Ordercited by 28
3 more
  1. 2021SynthBio: A Case Study in Human-AI Collaborative Curation of Text Datasetscited by 18
  2. 2020PHICON: Improving Generalization of Clinical Text De-identification Models via Data Augmentationcited by 14
  3. 2019Adversarial Learning of Privacy-Preserving Text Representations for De-Identification of Medical Recordscited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 0; 2024: 3; 2025: 4; 2026: 3.

11 papers.

  1. 2025A False Sense of Privacy: Evaluating Textual Data Sanitization Beyond Surface-level Privacy Leakagecited by 15
  2. 2026Large-scale online deanonymization with LLMscited by 13
  3. 2022Do CoNLL-2003 Named Entity Taggers Still Work Well in 2023?cited by 9
  4. 2024More RLHF, More Trust? On The Impact of Preference Alignment On Trustworthinesscited by 9
  5. 2024DIRI: Adversarial Patient Reidentification with Large Language Models for Evaluating Clinical Text Anonymizationcited by 4
  6. 2025Can Large Language Models Really Recognize Your Name?cited by 4
  7. 2025Towards Privacy-aware Mental Health AI Models: Advances, Challenges, and Opportunitiescited by 2
  8. 2024Comparing Feature-based and Context-aware Approaches to PII Generalization Level Predictioncited by 1
  9. 2026Beyond Indistinguishability: Measuring Extraction Risk in LLM APIscited by 1
  10. 2025Protecting De-identified Documents from Search-based Linkage Attackscited by 0
  11. 2026The Conundrum of Trustworthy Research on Attacking Personally Identifiable Information Removal Techniquescited by 0