Metraipolis › archive › dc_69408eeccc7d

Beyond Direct Identifiers: Probabilistic Privacy Risk Estimation for Privacy-Conscious LLM Query Delegation — arXiv:2608.09140

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.09140
License (as archived)
CC0-1.0
Archived
2026-10-19T08:50:55.171Z
SHA-256 of the archived text
54707e365c6b2a8f5618684e848374bf4baa409013f4bbf83ce3375aae9f0924
Stable URL of this page
https://metraipolis.com/documento/dc_69408eeccc7d

How to cite

Beyond Direct Identifiers: Probabilistic Privacy Risk Estimation for Privacy-Conscious LLM Query Delegation — arXiv:2608.09140 — arXiv (en), https://arxiv.org/abs/2608.09140 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_69408eeccc7d (Metraipolis archive dc_69408eeccc7d, sha256 54707e365c6b2a8f5618684e848374bf4baa409013f4bbf83ce3375aae9f0924)
BibTeX
@misc{dc69408eeccc7d,
  title = {Beyond Direct Identifiers: Probabilistic Privacy Risk Estimation for Privacy-Conscious LLM Query Delegation — arXiv:2608.09140},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.09140},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_69408eeccc7d; sha256 54707e365c6b2a8f5618684e848374bf4baa409013f4bbf83ce3375aae9f0924},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.09140. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Recent work on protecting privacy during user-LLM interactions often focuses on direct, explicit identifiers: the personally-identifiable information (PII) captured by standard detectors. One such approach is Privacy-Conscious Delegation (PCD), where a local LLM acts as an intermediary. However, privacy risk does not stem solely from explicit identifiers but also PII-free self-disclosures, leaving users identifiable through combinations of quasi-identifying traits. We investigate a probabilistic variant of PCD, where we augment its objectives with an LLM-driven probabilistic estimation of k-anonymity. To facilitate this, we first create the PUPA-SD dataset, which contains naturalistic user queries with self-disclosure. Our preliminary results indicate that optimizing PAPILLON on PUPA-SD improves quality on unseen conversations across a variety of local models and produces the best privacy-utility balance for Llama-3.2-3B, while smaller models struggle to jointly optimize quality and privacy. We propose k-anonymity as a useful auxiliary metric for tackling PCD.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-11 in cs.CL).
Categorie arXiv: Cryptography and Security (cs.CR) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2608.09140 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-19T08:49:42.000Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).