Paper · 2020 · Open access

Differentially Private Language Models Benefit from Public Pre-training

Gavin Kerrigan, Dylan Slack, Jens Tuyls

Proceedings of the Second Workshop on Privacy in NLP, 39–45 · Published 1 January 2020 · DOI 10.18653/v1/2020.privatenlp-1.5

Open PDFView at the publisherCite this paper

In brief

“Differentially Private Language Models Benefit from Public Pre-training” is a 2020 paper by Gavin Kerrigan and 2 co-authors, published in Proceedings of the Second Workshop on Privacy in NLP. OpenAlex counts 84 citations as of 19 November 2026. It is open access under a CC BY licence: the full text is free to read.

Citations
84
Most cited in
2025 (20)
Field-weighted impact
2.16
References
15
Authors
3

Abstract

Language modeling is a keystone task in natural language processing.When training a language model on sensitive information, differential privacy (DP) allows us to quantify the degree to which our private data is protected.However, training algorithms which enforce differential privacy often lead to degradation in model quality.We study the feasibility of learning a language model which is simultaneously high-quality and privacy preserving by tuning a public base model on a private corpus.We find that DP fine-tuning boosts the performance of language models in the private domain, making the training of such models possible.1

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Kerrigan, G., Slack, D., & Tuyls, J. (2020). Differentially Private Language Models Benefit from Public Pre-training. Proceedings of the Second Workshop on Privacy in NLP, 39–45. https://doi.org/10.18653/v1/2020.privatenlp-1.5

MLA 9

Kerrigan, Gavin, et al. “Differentially Private Language Models Benefit from Public Pre-training.” Proceedings of the Second Workshop on Privacy in NLP, 2020, pp. 39–45. https://doi.org/10.18653/v1/2020.privatenlp-1.5.

Chicago

Kerrigan, Gavin, Dylan Slack, and Jens Tuyls. 2020. “Differentially Private Language Models Benefit from Public Pre-training.” Proceedings of the Second Workshop on Privacy in NLP: 39–45. https://doi.org/10.18653/v1/2020.privatenlp-1.5.

Harvard

Kerrigan, G., Slack, D. and Tuyls, J. (2020) ‘Differentially Private Language Models Benefit from Public Pre-training’, Proceedings of the Second Workshop on Privacy in NLP, pp. 39–45. doi:10.18653/v1/2020.privatenlp-1.5.

IEEE

G. Kerrigan, D. Slack, and J. Tuyls, “Differentially Private Language Models Benefit from Public Pre-training,” Proceedings of the Second Workshop on Privacy in NLP, pp. 39–45, 2020, doi: 10.18653/v1/2020.privatenlp-1.5.

BibTeX

@article{kerrigan2020differentially,
  title     = {{Differentially Private Language Models Benefit from Public Pre-training}},
  author    = {Kerrigan, Gavin and Slack, Dylan and Tuyls, Jens},
  journal   = {Proceedings of the Second Workshop on Privacy in NLP},
  year      = {2020},
  pages     = {39--45},
  doi       = {10.18653/v1/2020.privatenlp-1.5},
  url       = {https://doi.org/10.18653/v1/2020.privatenlp-1.5}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Citations by year

YearCitationsShare of the best year
202011
20211010
20221111
20231414
20241717
20252020
20261111

Its field-weighted citation impact is 2.16: it has been cited 2.16 times as often as the average work of the same field, type and year (1.00 is the world average).

Topic and keywords

Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data

  • differential privacy
  • language models
  • fine-tuning
  • privacy-preserving machine learning
  • sensitive data

Work with this paper

Most-cited papers in Computer Science

All highly cited Computer Science papers

Questions about this paper

How do I cite this paper in APA style?

Kerrigan, G., Slack, D., & Tuyls, J. (2020). Differentially Private Language Models Benefit from Public Pre-training. Proceedings of the Second Workshop on Privacy in NLP, 39–45. https://doi.org/10.18653/v1/2020.privatenlp-1.5

How many times has this paper been cited?

84 times, according to OpenAlex on 19 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access: sign in and open the PDF from this page.

Who wrote this paper?

Gavin Kerrigan, Dylan Slack, Jens Tuyls.

What is the DOI of this paper?

10.18653/v1/2020.privatenlp-1.5, permanently at https://doi.org/10.18653/v1/2020.privatenlp-1.5. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.