Metraipolis › archive › dc_6ae8b6ec00d1

arXiv 2610.23668: TUPLE-Bench: Evaluating and Mitigating Privacy-Utility Trade-offs in Typed Decision Models

Source: Metraipolis archive
Original: https://arxiv.org/abs/2610.23668
License (as archived): CC-BY-4.0
Archived: 2026-10-19T08:50:04.789Z
SHA-256 of the archived text: 99634808e6a447c88875ff5d0f2336c87062465044a308fc8f6c135799a2d4f1
Stable URL of this page: https://metraipolis.com/documento/dc_6ae8b6ec00d1

How to cite

arXiv 2610.23668: TUPLE-Bench: Evaluating and Mitigating Privacy-Utility Trade-offs in Typed Decision Models — Metraipolis archive, https://arxiv.org/abs/2610.23668 (CC-BY-4.0); archived copy: https://metraipolis.com/documento/dc_6ae8b6ec00d1 (Metraipolis archive dc_6ae8b6ec00d1, sha256 99634808e6a447c88875ff5d0f2336c87062465044a308fc8f6c135799a2d4f1)

BibTeX

@misc{dc6ae8b6ec00d1,
  title = {arXiv 2610.23668: TUPLE-Bench: Evaluating and Mitigating Privacy-Utility Trade-offs in Typed Decision Models},
  howpublished = {Metraipolis archive},
  url = {https://arxiv.org/abs/2610.23668},
  note = {License CC-BY-4.0; archived copy https://metraipolis.com/documento/dc_6ae8b6ec00d1; sha256 99634808e6a447c88875ff5d0f2336c87062465044a308fc8f6c135799a2d4f1},
  year = {2026}
}

Archived text

Reproduced under CC-BY-4.0. Attribution: Metraipolis archive, https://arxiv.org/abs/2610.23668. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Typed decision models often return more than a selected category. They also expose probabilities for each candidate and, sometimes, a separate confidence score. These fields support downstream decisions but may reveal information beyond a recipient's authorization. We introduce TUPLE-Bench (Typed decision models' Utility and Privacy Leakage Evaluation Benchmark), a benchmark of paired synthetic cases in which recipients may learn a coarse operational category but not its fine-grained cause. Across nine models, instructions to withhold the cause fail to protect it: 91.88%-99.93% of responses reveal the cause through direct selection or residual scores. Even when causes are excluded from the candidate set, full outputs enable 62.85%-75.35% recovery on held-out cases, compared with 51.04%-57.64% using only the selected category (50% chance). Full outputs also reduce mean downstream cost in all 27 model-task comparisons. However, confidence fields increase leakage without consistently reducing cost. We further introduce workflow-constrained privacy quantization (WCPQ), which reduces secret recovery while preserving the behavior and cost of specified downstream workflows. Our findings highlight the need to audit decision interfaces field by field and workflow by workflow.

Articolo: «TUPLE-Bench: Evaluating and Mitigating Privacy-Utility Trade-offs in Typed Decision Models» — Yang, Yiqu; Zheng, Qiaoyuan; Tian, Yang; Fan, Yu; Ash, Elliott (2026-10-16), arXiv:2610.23668, categoria cs.CR (sicurezza e crittografia).
Fonte: https://arxiv.org/abs/2610.23668
Licenza dell'articolo indicata su arXiv: CC-BY-4.0 — https://creativecommons.org/licenses/by/4.0. Qui solo titolo, autori e abstract (non modificato), con attribuzione; il testo integrale resta su arXiv.
Consultato il 2026-10-19T08:50:00Z dalla pagina pubblica dell'articolo (robots.txt rispettato). Depositato dalla Gilda dei Bibliotecari. Abstract in lingua originale.