Conference paper · 2024 · Open access
Differentially Private Knowledge Distillation via Synthetic Text Generation
Findings of the Association for Computational Linguistics: ACL (Findings of ACL), 12957–12968 · Published 1 January 2024 · DOI 10.18653/v1/2024.findings-acl.769
In brief
“Differentially Private Knowledge Distillation via Synthetic Text Generation” is a 2024 conference paper by James Flemings and Murali Annavaram, published in Findings of the Association for Computational Linguistics: ACL (Findings of ACL). OpenAlex counts 29 citations as of 19 November 2026. It is open access under a CC BY licence: the full text is free to read at Findings of the Association for Computational Linguistics: ACL (Findings of ACL).
- Citations
- 29
- Most cited in
- 2025 (13)
- Field-weighted impact
- 2.42
- References
- 32
- Authors
- 2
Abstract
Large Language models (LLMs) are achieving state-of-the-art performance in many different downstream tasks.However, the increasing urgency of data privacy puts pressure on practitioners to train LLMs with Differential Privacy (DP) on private data.Concurrently, the exponential growth in parameter size of LLMs necessitates model compression before deployment of LLMs on resource-constrained devices or latency-sensitive applications.Differential privacy and model compression generally must trade off utility loss to achieve their objectives.Moreover, simultaneously applying both schemes can compound the utility degradation.To this end, we propose DistilDP: a novel differentially private knowledge distillation algorithm that exploits synthetic data generated by a differentially private teacher LLM.The knowledge of a teacher LLM is transferred onto the student in two ways: one way from the synthetic data itself-the hard labels, and the other way by the output distribution of the teacher evaluated on the synthetic data-the soft labels.Furthermore, if the teacher and student share a similar architectural structure, we can further distill knowledge by aligning the hidden representations between both.Our experimental results demonstrate that DistilDP can substantially improve the utility over existing baselines, at least 9.0 PPL on the Big Patent dataset, with strong privacy parameters, ϵ = 2.These promising results progress privacy-preserving compression of autoregressive LLMs.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Flemings, J., & Annavaram, M. (2024). Differentially Private Knowledge Distillation via Synthetic Text Generation. Findings of the Association for Computational Linguistics: ACL (Findings of ACL), 12957–12968. https://doi.org/10.18653/v1/2024.findings-acl.769
MLA 9
Flemings, James, and Murali Annavaram. “Differentially Private Knowledge Distillation via Synthetic Text Generation.” Findings of the Association for Computational Linguistics: ACL (Findings of ACL), 2024, pp. 12957–12968. https://doi.org/10.18653/v1/2024.findings-acl.769.
Chicago
Flemings, James and Murali Annavaram. 2024. “Differentially Private Knowledge Distillation via Synthetic Text Generation.” Findings of the Association for Computational Linguistics: ACL (Findings of ACL): 12957–12968. https://doi.org/10.18653/v1/2024.findings-acl.769.
Harvard
Flemings, J. and Annavaram, M. (2024) ‘Differentially Private Knowledge Distillation via Synthetic Text Generation’, Findings of the Association for Computational Linguistics: ACL (Findings of ACL), pp. 12957–12968. doi:10.18653/v1/2024.findings-acl.769.
IEEE
J. Flemings and M. Annavaram, “Differentially Private Knowledge Distillation via Synthetic Text Generation,” Findings of the Association for Computational Linguistics: ACL (Findings of ACL), pp. 12957–12968, 2024, doi: 10.18653/v1/2024.findings-acl.769.
BibTeX
@inproceedings{flemings2024differentially,
title = {{Differentially Private Knowledge Distillation via Synthetic Text Generation}},
author = {Flemings, James and Annavaram, Murali},
booktitle = {Findings of the Association for Computational Linguistics: ACL (Findings of ACL)},
year = {2024},
pages = {12957--12968},
doi = {10.18653/v1/2024.findings-acl.769},
url = {https://doi.org/10.18653/v1/2024.findings-acl.769}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Citations by year
| Year | Citations | Share of the best year |
|---|---|---|
| 2022 | 1 | |
| 2024 | 8 | |
| 2025 | 13 | |
| 2026 | 7 |
Its field-weighted citation impact is 2.42: it has been cited 2.42 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
- synthetic text generation
- large language models
- model compression
- soft labels
- autoregressive language models
Work with this paper
Most-cited papers in Computer Science
- ImageNet classification with deep convolutional neural networksAlex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton · Communications of the ACM · 2017 · 114,726 citations
- Long Short-Term MemorySepp Hochreiter, Jürgen Schmidhuber · Neural Computation · 1997 · 109,097 citations
- A Mathematical Theory of CommunicationClaude E. Shannon · Bell System Technical Journal · 1948 · 91,736 citations
- Deep learningYann LeCun, Yoshua Bengio, Geoffrey E. Hinton · Nature · 2015 · 90,226 citations
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksShaoqing Ren, Kaiming He, Ross Girshick and 1 more · IEEE Transactions on Pattern Analysis and Machine Intelligence · 2016 · 74,143 citations
Questions about this paper
How do I cite this paper in APA style?
Flemings, J., & Annavaram, M. (2024). Differentially Private Knowledge Distillation via Synthetic Text Generation. Findings of the Association for Computational Linguistics: ACL (Findings of ACL), 12957–12968. https://doi.org/10.18653/v1/2024.findings-acl.769
How many times has this paper been cited?
29 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 Findings of the Association for Computational Linguistics: ACL (Findings of ACL): sign in and open the PDF from this page.
Who wrote this paper?
James Flemings, Murali Annavaram.
What is the DOI of this paper?
10.18653/v1/2024.findings-acl.769, permanently at https://doi.org/10.18653/v1/2024.findings-acl.769. 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.