Metraipolis › archive › dc_8766743ea7ba

QuanText: Protecting Dataset-Level Secrets in Textual Data Sharing — arXiv:2609.17995

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.17995
License (as archived)
CC0-1.0
Archived
2026-10-15T14:31:49.553Z
SHA-256 of the archived text
cf45afaf0ea950fa9d59eb0e47da73820cdd5d42c94bc8bcc9ea127b0fa1f562
Stable URL of this page
https://metraipolis.com/documento/dc_8766743ea7ba

How to cite

QuanText: Protecting Dataset-Level Secrets in Textual Data Sharing — arXiv:2609.17995 — arXiv (en), https://arxiv.org/abs/2609.17995 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_8766743ea7ba (Metraipolis archive dc_8766743ea7ba, sha256 cf45afaf0ea950fa9d59eb0e47da73820cdd5d42c94bc8bcc9ea127b0fa1f562)
BibTeX
@misc{dc8766743ea7ba,
  title = {QuanText: Protecting Dataset-Level Secrets in Textual Data Sharing — arXiv:2609.17995},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.17995},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_8766743ea7ba; sha256 cf45afaf0ea950fa9d59eb0e47da73820cdd5d42c94bc8bcc9ea127b0fa1f562},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.17995. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Natural-language datasets support many downstream applications and research studies, but releasing text can reveal sensitive global properties of the underlying data source, such as the proportion of records associated with a particular gender, diagnosis, or political stance. Existing work has largely focused on property inference attacks that recover such global properties, while defenses for protecting these dataset-level secrets remain limited. Differential privacy, although effective for protecting individual records, provides only weak protection for aggregate properties. We propose Randomized Quantization for Text (QuanText), a training-free and large-language-model-agnostic data release mechanism that protects global secrets in textual datasets while preserving data utility. Given a dataset-level secret, such as the proportion of records with a particular diagnosis, and attributes whose utility should be preserved, such as topic and sentiment, QuanText perturbs both the secret distribution and the distributions of correlated attributes. It does so by constructing candidate release distributions over secret and non-secret attributes, randomly selecting a candidate sufficiently close to the private empirical distribution, and rewriting each private text sample to match the selected distribution using attribute-related snippets from the original text. QuanText is inspired by the Statistic Maximal Leakage (SML) framework, which bounds leakage about a secret function of a data distribution. Under idealized conditions, we show that QuanText satisfies an SML guarantee. Since these conditions may not hold exactly in practice, we also evaluate QuanText empirically on real-world datasets. Our results show that QuanText achieves a better empirical privacy-utility trade-off than competing data generation baselines.

Tema degli esploratori: fiducia digitale e crittografia (retro-catalogo arXiv: annunci del 2026-09-17 in cs.CR).
Categorie arXiv: Machine Learning (cs.LG) ; Cryptography and Security (cs.CR).
Pagina del lavoro: https://arxiv.org/abs/2609.17995 — 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 2026-10-15T13:56:52.896Z dalla pagina https://arxiv.org/catchup/cs.CR/2026-09-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).