Conference paper · 2025 · Open access

Differentially Private Learning Needs Better Model Initialization and Self-Distillation

Ivoline C. Ngong, Joseph P. Near, Niloofar Mireshghallah

Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT), 9009–9027 · Published 1 January 2025 · DOI 10.18653/v1/2025.naacl-long.455

Open PDFView at the publisherCite this paper

In brief

“Differentially Private Learning Needs Better Model Initialization and Self-Distillation” is a 2025 conference paper by Ivoline C. Ngong and 2 co-authors, published in Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT). OpenAlex counts 4 citations as of 19 November 2026. It is open access under a CC BY licence: the full text is free to read at Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT).

Citations
4
Most cited in
2026 (3)
Field-weighted impact
0.83
References
47
Authors
3

Abstract

Ivoline C. Ngong, Joseph Near, Niloofar Mireshghallah. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

Cite this paper

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

APA 7

Ngong, I. C., Near, J. P., & Mireshghallah, N. (2025). Differentially Private Learning Needs Better Model Initialization and Self-Distillation. Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT), 9009–9027. https://doi.org/10.18653/v1/2025.naacl-long.455

MLA 9

Ngong, Ivoline C., et al. “Differentially Private Learning Needs Better Model Initialization and Self-Distillation.” Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT), 2025, pp. 9009–9027. https://doi.org/10.18653/v1/2025.naacl-long.455.

Chicago

Ngong, Ivoline C., Joseph P. Near, and Niloofar Mireshghallah. 2025. “Differentially Private Learning Needs Better Model Initialization and Self-Distillation.” Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT): 9009–9027. https://doi.org/10.18653/v1/2025.naacl-long.455.

Harvard

Ngong, I. C., Near, J. P. and Mireshghallah, N. (2025) ‘Differentially Private Learning Needs Better Model Initialization and Self-Distillation’, Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT), pp. 9009–9027. doi:10.18653/v1/2025.naacl-long.455.

IEEE

I. C. Ngong, J. P. Near, and N. Mireshghallah, “Differentially Private Learning Needs Better Model Initialization and Self-Distillation,” Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT), pp. 9009–9027, 2025, doi: 10.18653/v1/2025.naacl-long.455.

BibTeX

@inproceedings{ngong2025differentially,
  title     = {{Differentially Private Learning Needs Better Model Initialization and Self-Distillation}},
  author    = {Ngong, Ivoline C. and Near, Joseph P. and Mireshghallah, Niloofar},
  booktitle = {Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT)},
  year      = {2025},
  pages     = {9009--9027},
  doi       = {10.18653/v1/2025.naacl-long.455},
  url       = {https://doi.org/10.18653/v1/2025.naacl-long.455}
}

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

Topic and keywords

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

  • differential privacy
  • model initialization
  • self-distillation
  • knowledge distillation
  • privacy-preserving machine learning
  • neural network training

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?

Ngong, I. C., Near, J. P., & Mireshghallah, N. (2025). Differentially Private Learning Needs Better Model Initialization and Self-Distillation. Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT), 9009–9027. https://doi.org/10.18653/v1/2025.naacl-long.455

How many times has this paper been cited?

4 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 at Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT): sign in and open the PDF from this page.

Who wrote this paper?

Ivoline C. Ngong, Joseph P. Near, Niloofar Mireshghallah.

What is the DOI of this paper?

10.18653/v1/2025.naacl-long.455, permanently at https://doi.org/10.18653/v1/2025.naacl-long.455. 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.