Conference paper · 2025 · Open access

PIG: Privacy Jailbreak Attack on LLMs via Gradient-based Iterative In-Context Optimization

Yidan Wang, Yanan Cao, Yubing Ren, Fang Fang, Zheng Lin, Binxing Fang

Proceedings of the conference - Association for Computational Linguistics. Meeting, 9645–9660 · Published 1 January 2025 · DOI 10.18653/v1/2025.acl-long.475

Open PDFView at the publisherCite this paper

In brief

“PIG: Privacy Jailbreak Attack on LLMs via Gradient-based Iterative In-Context Optimization” is a 2025 conference paper by Yidan Wang and 5 co-authors, published in Proceedings of the conference - Association for Computational Linguistics. Meeting. OpenAlex counts 16 citations as of 19 November 2026, which puts it in the top 4% of works published in 2025 by citations. It is open access under a CC BY licence: the full text is free to read at Proceedings of the conference - Association for Computational Linguistics. Meeting.

Citations
16
Most cited in
2026 (12)
Field-weighted impact
3.33
References
35
Authors
6

Abstract

Large Language Models (LLMs) excel in various domains but pose inherent privacy risks.Existing methods to evaluate privacy leakage in LLMs often use memorized prefixes or simple instructions to extract data, both of which well-alignment models can easily block.Meanwhile, Jailbreak attacks bypass LLM safety mechanisms to generate harmful content, but their role in privacy scenarios remains underexplored.In this paper, we examine the effectiveness of jailbreak attacks in extracting sensitive information, bridging privacy leakage and jailbreak attacks in LLMs.Moreover, we propose PIG, a novel framework targeting Personally Identifiable Information (PII) and addressing the limitations of current jailbreak methods.Specifically, PIG identifies PII entities and their types in privacy queries, uses in-context learning to build a privacy context, and iteratively updates it with three gradient-based strategies to elicit target PII.We evaluate PIG and existing jailbreak methods using two privacy-related datasets.Experiments on four white-box and two blackbox LLMs show that PIG outperforms baseline methods and achieves state-of-the-art (SoTA) results.The results underscore significant privacy risks in LLMs, emphasizing the need for stronger safeguards.

Cite this paper

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

APA 7

Wang, Y., Cao, Y., Ren, Y., Fang, F., Lin, Z., & Fang, B. (2025). PIG: Privacy Jailbreak Attack on LLMs via Gradient-based Iterative In-Context Optimization. Proceedings of the conference - Association for Computational Linguistics. Meeting, 9645–9660. https://doi.org/10.18653/v1/2025.acl-long.475

MLA 9

Wang, Yidan, et al. “PIG: Privacy Jailbreak Attack on LLMs via Gradient-based Iterative In-Context Optimization.” Proceedings of the conference - Association for Computational Linguistics. Meeting, 2025, pp. 9645–9660. https://doi.org/10.18653/v1/2025.acl-long.475.

Chicago

Wang, Yidan, Yanan Cao, Yubing Ren, Fang Fang, Zheng Lin, and Binxing Fang. 2025. “PIG: Privacy Jailbreak Attack on LLMs via Gradient-based Iterative In-Context Optimization.” Proceedings of the conference - Association for Computational Linguistics. Meeting: 9645–9660. https://doi.org/10.18653/v1/2025.acl-long.475.

Harvard

Wang, Y. et al. (2025) ‘PIG: Privacy Jailbreak Attack on LLMs via Gradient-based Iterative In-Context Optimization’, Proceedings of the conference - Association for Computational Linguistics. Meeting, pp. 9645–9660. doi:10.18653/v1/2025.acl-long.475.

IEEE

Y. Wang, Y. Cao, Y. Ren, F. Fang, Z. Lin, and B. Fang, “PIG: Privacy Jailbreak Attack on LLMs via Gradient-based Iterative In-Context Optimization,” Proceedings of the conference - Association for Computational Linguistics. Meeting, pp. 9645–9660, 2025, doi: 10.18653/v1/2025.acl-long.475.

BibTeX

@inproceedings{wang2025privacy,
  title     = {{PIG: Privacy Jailbreak Attack on LLMs via Gradient-based Iterative In-Context Optimization}},
  author    = {Wang, Yidan and Cao, Yanan and Ren, Yubing and Fang, Fang and Lin, Zheng and Fang, Binxing},
  booktitle = {Proceedings of the conference - Association for Computational Linguistics. Meeting},
  year      = {2025},
  pages     = {9645--9660},
  doi       = {10.18653/v1/2025.acl-long.475},
  url       = {https://doi.org/10.18653/v1/2025.acl-long.475}
}

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

Topic and keywords

Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning

  • jailbreak attacks
  • large language models
  • privacy leakage
  • personally identifiable information
  • in-context learning

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?

Wang, Y., Cao, Y., Ren, Y., Fang, F., Lin, Z., & Fang, B. (2025). PIG: Privacy Jailbreak Attack on LLMs via Gradient-based Iterative In-Context Optimization. Proceedings of the conference - Association for Computational Linguistics. Meeting, 9645–9660. https://doi.org/10.18653/v1/2025.acl-long.475

How many times has this paper been cited?

16 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 Proceedings of the conference - Association for Computational Linguistics. Meeting: sign in and open the PDF from this page.

Who wrote this paper?

Yidan Wang, Yanan Cao, Yubing Ren, Fang Fang, Zheng Lin, Binxing Fang.

What is the DOI of this paper?

10.18653/v1/2025.acl-long.475, permanently at https://doi.org/10.18653/v1/2025.acl-long.475. 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.