Conference paper · 2026 · Open access

Reconstruction Attack-Resistant Inference Paradigm for LLM Cloud Services

Zipeng Ye, Wenjian Luo, Qi Zhou, Yubo Tang

Proceedings of the AAAI Conference on Artificial Intelligence, 40(21), 17939–17947 · Published 14 March 2026 · DOI 10.1609/aaai.v40i21.38853

Open PDFView at the publisherCite this paper

In brief

“Reconstruction Attack-Resistant Inference Paradigm for LLM Cloud Services” is a 2026 conference paper by Zipeng Ye and 3 co-authors, published in Proceedings of the AAAI Conference on Artificial Intelligence. OpenAlex counts 2 citations as of 19 November 2026, which puts it in the top 6% of works published in 2026 by citations. It is open access: the full text is free to read at Proceedings of the AAAI Conference on Artificial Intelligence.

Citations
2
Most cited in
2026 (2)
Field-weighted impact
1.77
References
29
Authors
4

Abstract

Large language models (LLMs) have seen remarkable growth in recent years. To leverage convenient LLM cloud services, users are inevitably to upload their prompts. Additionally, for tasks such as translation, reading comprehension, and summarization, associated files or context are inherently needed, whether or not they contain user privacy information. Despite the rapid progress in LLM capabilities, research on preserving user privacy during inference has been relatively scarce. To this end, this paper conducts some exploratory research in this domain. Firstly, we show that (1) the embedding space of tokens is highly sparse, and (2) LLMs primarily function in the orthogonal subspace of embedding space, these two factors making privacy extremely vulnerable. Then, we analyze the structural characteristics of LLMs and design a distributed privacy-preserving inference paradigm which can effectively resist privacy attacks. Finally, we perform a thorough evaluation of the defended models on mainstream tasks and find that low-bit quantization techniques can be effectively combined with our inference paradigm, achieving a balance between privacy, utility, and runtime memory efficiency.

Cite this paper

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

APA 7

Ye, Z., Luo, W., Zhou, Q., & Tang, Y. (2026). Reconstruction Attack-Resistant Inference Paradigm for LLM Cloud Services. Proceedings of the AAAI Conference on Artificial Intelligence, 40(21), 17939–17947. https://doi.org/10.1609/aaai.v40i21.38853

MLA 9

Ye, Zipeng, et al. “Reconstruction Attack-Resistant Inference Paradigm for LLM Cloud Services.” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 40, no. 21, 2026, pp. 17939–17947. https://doi.org/10.1609/aaai.v40i21.38853.

Chicago

Ye, Zipeng, Wenjian Luo, Qi Zhou, and Yubo Tang. 2026. “Reconstruction Attack-Resistant Inference Paradigm for LLM Cloud Services.” Proceedings of the AAAI Conference on Artificial Intelligence 40 (21): 17939–17947. https://doi.org/10.1609/aaai.v40i21.38853.

Harvard

Ye, Z. et al. (2026) ‘Reconstruction Attack-Resistant Inference Paradigm for LLM Cloud Services’, Proceedings of the AAAI Conference on Artificial Intelligence, 40(21), pp. 17939–17947. doi:10.1609/aaai.v40i21.38853.

IEEE

Z. Ye, W. Luo, Q. Zhou, and Y. Tang, “Reconstruction Attack-Resistant Inference Paradigm for LLM Cloud Services,” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 40, no. 21, pp. 17939–17947, 2026, doi: 10.1609/aaai.v40i21.38853.

BibTeX

@inproceedings{ye2026reconstruction,
  title     = {{Reconstruction Attack-Resistant Inference Paradigm for LLM Cloud Services}},
  author    = {Ye, Zipeng and Luo, Wenjian and Zhou, Qi and Tang, Yubo},
  booktitle = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year      = {2026},
  volume    = {40},
  number    = {21},
  pages     = {17939--17947},
  doi       = {10.1609/aaai.v40i21.38853},
  url       = {https://doi.org/10.1609/aaai.v40i21.38853}
}

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

  • large language models
  • privacy-preserving inference
  • reconstruction attacks
  • orthogonal subspaces
  • distributed inference
  • low-bit quantization
  • user privacy
  • memory 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?

Ye, Z., Luo, W., Zhou, Q., & Tang, Y. (2026). Reconstruction Attack-Resistant Inference Paradigm for LLM Cloud Services. Proceedings of the AAAI Conference on Artificial Intelligence, 40(21), 17939–17947. https://doi.org/10.1609/aaai.v40i21.38853

How many times has this paper been cited?

2 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 AAAI Conference on Artificial Intelligence: sign in and open the PDF from this page.

Who wrote this paper?

Zipeng Ye, Wenjian Luo, Qi Zhou, Yubo Tang.

What is the DOI of this paper?

10.1609/aaai.v40i21.38853, permanently at https://doi.org/10.1609/aaai.v40i21.38853. 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.