# Token-Level Privacy in Large Language Models

arXiv: https://arxiv.org/abs/2503.03652 · HTML: https://mlsanity.com/paper/2503.03652/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 35 references that are in the map.
- First posted on arXiv 5 Mar 2025.

## Builds on (most cited first, 18 of 18)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://mlsanity.com/paper/1804.07461/) (2018)
- [Scaling Instruction-Finetuned Language Models](https://mlsanity.com/paper/2210.11416/) (2022)
- [Extracting Training Data from Large Language Models](https://mlsanity.com/paper/2012.07805/) (2020)
- [Qwen2 Technical Report](https://mlsanity.com/paper/2407.10671/) (2024)
- [SWAG: A Large-Scale Adversarial Dataset for Grounded Commonsense Inference](https://mlsanity.com/paper/1808.05326/) (2018)
- [Privacy- and Utility-Preserving Textual Analysis via Calibrated Multivariate Perturbations](https://mlsanity.com/paper/1910.08902/) (2019)
- [Differential Privacy for Text Analytics via Natural Text Sanitization](https://mlsanity.com/paper/2106.01221/) (2021)
- [Differentially Private Representation for NLP: Formal Guarantee and An Empirical Study on Privacy and Fairness](https://mlsanity.com/paper/2010.01285/) (2020)
- [Natural Language Understanding with Privacy-Preserving BERT](https://mlsanity.com/paper/2104.07504/) (2021)
- [Generalised Differential Privacy for Text Document Processing](https://mlsanity.com/paper/1811.10256/) (2018)
- [A Customized Text Sanitization Mechanism with Differential Privacy](https://mlsanity.com/paper/2207.01193/) (2022)
- [Local Differential Privacy for Deep Learning](https://mlsanity.com/paper/1908.02997/) (2019)
- [Audio De-identification: A New Entity Recognition Task](https://mlsanity.com/paper/1903.07037/) (2019)
- [Adversarial Learning of Privacy-Preserving Text Representations for De-Identification of Medical Records](https://mlsanity.com/paper/1906.05000/) (2019)
- [Protecting Privacy in Classifiers by Token Manipulation](https://mlsanity.com/paper/2407.01334/) (2024)

## Abstract (arXiv)

The use of language models as remote services requires transmitting private information to external providers, raising significant privacy concerns. This process not only risks exposing sensitive data to untrusted service providers but also leaves it vulnerable to interception by eavesdroppers. Existing privacy-preserving methods for natural language processing (NLP) interactions primarily rely on semantic similarity, overlooking the role of contextual information. In this work, we introduce dchi-stencil, a novel token-level privacy-preserving mechanism that integrates contextual and semantic information while ensuring strong privacy guarantees under the dchi differential privacy framework, achieving 2epsilon-dchi-privacy. By incorporating both semantic and contextual nuances, dchi-stencil achieves a robust balance between privacy and utility. We evaluate dchi-stencil using state-of-the-art language models and diverse datasets, achieving comparable and even better trade-off between utility and privacy compared to existing methods. This work highlights the potential of dchi-stencil to set a new standard for privacy-preserving NLP in modern, high-risk applications.