Metraipolis › archive › dc_efd196b7a14d

SoK: Privacy Attacks on Machine Learning via Explainable AI — arXiv:2609.10627

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.10627
License (as archived)
CC0-1.0
Archived
2026-10-16T05:46:20.305Z
SHA-256 of the archived text
c8321f6b7de5f23400423bc45da85b9d89d729deaf0e1f00a79901cf3bf29530
Stable URL of this page
https://metraipolis.com/documento/dc_efd196b7a14d

How to cite

SoK: Privacy Attacks on Machine Learning via Explainable AI — arXiv:2609.10627 — arXiv (en), https://arxiv.org/abs/2609.10627 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_efd196b7a14d (Metraipolis archive dc_efd196b7a14d, sha256 c8321f6b7de5f23400423bc45da85b9d89d729deaf0e1f00a79901cf3bf29530)
BibTeX
@misc{dcefd196b7a14d,
  title = {SoK: Privacy Attacks on Machine Learning via Explainable AI — arXiv:2609.10627},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.10627},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_efd196b7a14d; sha256 c8321f6b7de5f23400423bc45da85b9d89d729deaf0e1f00a79901cf3bf29530},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.10627. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Machine learning explanations reveal model behavior beyond predictions, creating attack surfaces for model confidentiality and data privacy. We systematize 25 studies that exploit explanations for model extraction, membership inference, and model inversion, treating attribute inference as partial inversion. Existing work is often labeled only black- or white-box, obscuring substantial differences in what explanation signal reaches an adversary. We therefore separate model knowledge from explanation acquisition and identify five paths: target-released, attacker-derived, secondary disclosure, privileged access, and released global artifacts. Across these paths, explanations reduce extraction cost, expose membership signals through explanation statistics, recourse distance, and explanation-guided robustness, and support spatial or algebraic reconstruction of private inputs. We compare system and threat models, explanation signals, auxiliary knowledge, target models, modalities, query budgets, evaluation metrics, reported performance, and defenses. Our analysis shows that no explanation family is uniformly unsafe and no defense is uniformly effective. Risk depends on which signal is exposed, how it is acquired, which asset is targeted, and what the attacker already knows. We argue that explanation privacy should therefore be evaluated as an end-to-end disclosure problem, with defenses matched to the acquisition path and protected asset.

Tema degli esploratori: fiducia digitale e crittografia (retro-catalogo arXiv: annunci del 2026-09-11 in cs.CR).
Categorie arXiv: Cryptography and Security (cs.CR) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.10627 — 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 2026-10-16T05:30:39.371Z dalla pagina https://arxiv.org/catchup/cs.CR/2026-09-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).