Preprint · 2025 · Open access

LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models

Delong Ran, Xinlei He, Tianshuo Cong, Anyu Wang, Qi Li, Wang, Xiaoyun

arXiv (Cornell University) · Published 24 July 2025 · DOI 10.48550/arxiv.2507.18302

Open PDFView at the publisherCite this paper

In brief

“LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models” is a 2025 preprint by Delong Ran and 5 co-authors, published in arXiv (Cornell University). OpenAlex counts 15 citations as of 19 November 2026. It is open access under a CC BY licence: the full text is free to read at arXiv (Cornell University).

Citations
15
Most cited in
2026 (14)
References
37
Authors
6

Abstract

Language Models (LMs) typically adhere to a "pre-training and fine-tuning" paradigm, where a universal pre-trained model can be fine-tuned to cater to various specialized domains. Low-Rank Adaptation (LoRA) has gained the most widespread use in LM fine-tuning due to its lightweight computational cost and remarkable performance. Because the proportion of parameters tuned by LoRA is relatively small, there might be a misleading impression that the LoRA fine-tuning data is invulnerable to Membership Inference Attacks (MIAs). However, we identify that utilizing the pre-trained model can induce more information leakage, which is neglected by existing MIAs. Therefore, we introduce LoRA-Leak, a holistic evaluation framework for MIAs against the fine-tuning datasets of LMs. LoRA-Leak incorporates fifteen membership inference attacks, including ten existing MIAs, and five improved MIAs that leverage the pre-trained model as a reference. In experiments, we apply LoRA-Leak to three advanced LMs across three popular natural language processing tasks, demonstrating that LoRA-based fine-tuned LMs are still vulnerable to MIAs (e.g., 0.775 AUC under conservative fine-tuning settings). We also applied LoRA-Leak to different fine-tuning settings to understand the resulting privacy risks. We further explore four defenses and find that only dropout and excluding specific LM layers during fine-tuning effectively mitigate MIA risks while maintaining utility. We highlight that under the "pre-training and fine-tuning" paradigm, the existence of the pre-trained model makes MIA a more severe risk for LoRA-based LMs. We hope that our findings can provide guidance on data privacy protection for specialized LM providers.

Cite this paper

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

APA 7

Ran, D., He, X., Cong, T., Wang, A., Li, Q., & Xiaoyun, W. (2025). LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2507.18302

MLA 9

Ran, Delong, et al. “LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models.” arXiv (Cornell University), 2025. https://doi.org/10.48550/arxiv.2507.18302.

Chicago

Ran, Delong, Xinlei He, Tianshuo Cong, Anyu Wang, Qi Li, and Wang, Xiaoyun. 2025. “LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2507.18302.

Harvard

Ran, D. et al. (2025) ‘LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models’, arXiv (Cornell University). doi:10.48550/arxiv.2507.18302.

IEEE

D. Ran, X. He, T. Cong, A. Wang, Q. Li, and W. Xiaoyun, “LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models,” arXiv (Cornell University), 2025, doi: 10.48550/arxiv.2507.18302.

BibTeX

@misc{ran2025lora,
  title     = {{LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models}},
  author    = {Ran, Delong and He, Xinlei and Cong, Tianshuo and Wang, Anyu and Li, Qi and Xiaoyun, Wang,},
  howpublished = {arXiv (Cornell University)},
  year      = {2025},
  doi       = {10.48550/arxiv.2507.18302},
  url       = {https://doi.org/10.48550/arxiv.2507.18302}
}

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

  • membership inference attacks
  • LoRa
  • low-rank adaptation
  • language model fine-tuning
  • data privacy
  • information leakage
  • pre-trained models
  • privacy defenses

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?

Ran, D., He, X., Cong, T., Wang, A., Li, Q., & Xiaoyun, W. (2025). LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2507.18302

How many times has this paper been cited?

15 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?

Delong Ran, Xinlei He, Tianshuo Cong, Anyu Wang, Qi Li, Wang, Xiaoyun.

What is the DOI of this paper?

10.48550/arxiv.2507.18302, permanently at https://doi.org/10.48550/arxiv.2507.18302. 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.