Preprint · 2026 · Open access

OCELOT: Inference-Leakage Budgets for Privacy-Preserving LLM Agents

Jin Xie, Songze Li

arXiv (Cornell University) · Published 10 June 2026 · DOI 10.48550/arxiv.2606.12341

Open PDFView at the publisherCite this paper

In brief

“OCELOT: Inference-Leakage Budgets for Privacy-Preserving LLM Agents” is a 2026 preprint by Jin Xie and Songze Li, published in arXiv (Cornell University). OpenAlex counts 4 citations as of 19 November 2026. It is open access under a CC BY-NC-SA licence: the full text is free to read at arXiv (Cornell University).

Citations
4
Most cited in
2026 (4)
References
28
Authors
2

Abstract

Large language model (LLM) agents increasingly act on a user's behalf -- reading personal files, calling tools, transacting with external services -- possibly leaking personally identifiable information (PII) across trust boundaries at every step. Privacy here is a property not of a single output but of an entire trajectory, and three properties make it hard: leakage is cumulative, as individually innocuous releases accumulate across honest-but-curious or colluding sinks into inferences about a protected secret; bidirectional, as a malicious observation can inject instructions that turn the agent's own reasoning model against the user; and task-dependent, as the same field is necessary for one recipient yet gratuitous for another. Per-release contextual-integrity filters, information-flow controls, and posterior-leakage monitors each address part of this but none controls cumulative, inference-based leakage at runtime. We recast agent privacy as \emph{posterior-risk control} and present OCELOT, a runtime mediator that budgets how much an adversary's belief about a secret may improve across a trajectory, rather than filtering outputs. Its mechanism, \emph{Witness-Verified Declassification}, separates judgment from trust: an untrusted, locally fine-tuned defender model inspects each candidate release and emits structured evidence -- labeled atoms and proposed declassification operators -- which a deterministic verifier audits, charging a certified min-entropy cost for the chosen variant and authorizing the least-disclosing useful release under a sink-trust-weighted budget recorded on a tamper-evident ledger. Across diverse agent benchmarks and recent defenses, OCELOT attains significantly lower leakage at higher task utility, resists adaptive injection, jailbreak, cumulative inference, and sink collusion, and adds only modest overhead.

Cite this paper

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

APA 7

Xie, J., & Li, S. (2026). OCELOT: Inference-Leakage Budgets for Privacy-Preserving LLM Agents. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2606.12341

MLA 9

Xie, Jin, and Songze Li. “OCELOT: Inference-Leakage Budgets for Privacy-Preserving LLM Agents.” arXiv (Cornell University), 2026. https://doi.org/10.48550/arxiv.2606.12341.

Chicago

Xie, Jin and Songze Li. 2026. “OCELOT: Inference-Leakage Budgets for Privacy-Preserving LLM Agents.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2606.12341.

Harvard

Xie, J. and Li, S. (2026) ‘OCELOT: Inference-Leakage Budgets for Privacy-Preserving LLM Agents’, arXiv (Cornell University). doi:10.48550/arxiv.2606.12341.

IEEE

J. Xie and S. Li, “OCELOT: Inference-Leakage Budgets for Privacy-Preserving LLM Agents,” arXiv (Cornell University), 2026, doi: 10.48550/arxiv.2606.12341.

BibTeX

@misc{xie2026ocelot,
  title     = {{OCELOT: Inference-Leakage Budgets for Privacy-Preserving LLM Agents}},
  author    = {Xie, Jin and Li, Songze},
  howpublished = {arXiv (Cornell University)},
  year      = {2026},
  doi       = {10.48550/arxiv.2606.12341},
  url       = {https://doi.org/10.48550/arxiv.2606.12341}
}

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

  • LLM agents
  • privacy leakage
  • personally identifiable information
  • contextual integrity
  • jailbreak attacks

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?

Xie, J., & Li, S. (2026). OCELOT: Inference-Leakage Budgets for Privacy-Preserving LLM Agents. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2606.12341

How many times has this paper been cited?

4 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?

Jin Xie, Songze Li.

What is the DOI of this paper?

10.48550/arxiv.2606.12341, permanently at https://doi.org/10.48550/arxiv.2606.12341. 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.