Paper · 2026 · Open access

Adaptive Token-Weighted Differential Privacy for LLMs: Not All Tokens Require Equal Protection

Manjiang Yu, Priyanka Singh, Xue Li, Yang Cao

Lecture notes in computer science, 456–473 · Published 1 May 2026 · DOI 10.1007/978-3-032-16092-8_25

Open PDFView at the publisherCite this paper

In brief

“Adaptive Token-Weighted Differential Privacy for LLMs: Not All Tokens Require Equal Protection” is a 2026 paper by Manjiang Yu and 3 co-authors, published in Lecture notes in computer science. OpenAlex counts 3 citations as of 19 November 2026, which puts it in the top 4% of works published in 2026 by citations. It is open access: the full text is free to read at arXiv (Cornell University).

Citations
3
Most cited in
2026 (3)
Field-weighted impact
2.66
References
23
Authors
4

Cite this paper

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

APA 7

Yu, M., Singh, P., Li, X., & Cao, Y. (2026). Adaptive Token-Weighted Differential Privacy for LLMs: Not All Tokens Require Equal Protection. Lecture notes in computer science, 456–473. https://doi.org/10.1007/978-3-032-16092-8_25

MLA 9

Yu, Manjiang, et al. “Adaptive Token-Weighted Differential Privacy for LLMs: Not All Tokens Require Equal Protection.” Lecture notes in computer science, 2026, pp. 456–473. https://doi.org/10.1007/978-3-032-16092-8_25.

Chicago

Yu, Manjiang, Priyanka Singh, Xue Li, and Yang Cao. 2026. “Adaptive Token-Weighted Differential Privacy for LLMs: Not All Tokens Require Equal Protection.” Lecture notes in computer science: 456–473. https://doi.org/10.1007/978-3-032-16092-8_25.

Harvard

Yu, M. et al. (2026) ‘Adaptive Token-Weighted Differential Privacy for LLMs: Not All Tokens Require Equal Protection’, Lecture notes in computer science, pp. 456–473. doi:10.1007/978-3-032-16092-8_25.

IEEE

M. Yu, P. Singh, X. Li, and Y. Cao, “Adaptive Token-Weighted Differential Privacy for LLMs: Not All Tokens Require Equal Protection,” Lecture notes in computer science, pp. 456–473, 2026, doi: 10.1007/978-3-032-16092-8_25.

BibTeX

@article{yu2026adaptive,
  title     = {{Adaptive Token-Weighted Differential Privacy for LLMs: Not All Tokens Require Equal Protection}},
  author    = {Yu, Manjiang and Singh, Priyanka and Li, Xue and Cao, Yang},
  journal   = {Lecture notes in computer science},
  year      = {2026},
  pages     = {456--473},
  doi       = {10.1007/978-3-032-16092-8_25},
  url       = {https://doi.org/10.1007/978-3-032-16092-8_25}
}

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
  • large language models
  • DP-SGD
  • privacy-preserving fine-tuning
  • training efficiency

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?

Yu, M., Singh, P., Li, X., & Cao, Y. (2026). Adaptive Token-Weighted Differential Privacy for LLMs: Not All Tokens Require Equal Protection. Lecture notes in computer science, 456–473. https://doi.org/10.1007/978-3-032-16092-8_25

How many times has this paper been cited?

3 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 arXiv (Cornell University): sign in and open the PDF from this page.

Who wrote this paper?

Manjiang Yu, Priyanka Singh, Xue Li, Yang Cao.

What is the DOI of this paper?

10.1007/978-3-032-16092-8_25, permanently at https://doi.org/10.1007/978-3-032-16092-8_25. 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.