Metraipolis › archive › dc_6fad9e6c8aed
Does the Readout Bypass Leak the Input? A Feature-Visibility Audit of Hybrid Quantum-Classical Models — arXiv:2609.38720
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.38720
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T13:43:23.382Z
- SHA-256 of the archived text
377a1cf9133ceb11b5b33d21e339e76f286c08b44c65472b564aa8fba6863218- Stable URL of this page
https://metraipolis.com/documento/dc_6fad9e6c8aed
How to cite
Does the Readout Bypass Leak the Input? A Feature-Visibility Audit of Hybrid Quantum-Classical Models — arXiv:2609.38720 — arXiv (en), https://arxiv.org/abs/2609.38720 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6fad9e6c8aed (Metraipolis archive dc_6fad9e6c8aed, sha256 377a1cf9133ceb11b5b33d21e339e76f286c08b44c65472b564aa8fba6863218)
BibTeX
@misc{dc6fad9e6c8aed,
title = {Does the Readout Bypass Leak the Input? A Feature-Visibility Audit of Hybrid Quantum-Classical Models — arXiv:2609.38720},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.38720},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6fad9e6c8aed; sha256 377a1cf9133ceb11b5b33d21e339e76f286c08b44c65472b564aa8fba6863218},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.38720. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Readout-side residual hybrids concatenate raw inputs with measured quantum features. Under single-example gradient sharing, a biased first linear layer admits standard analytic recovery of its input, so the bypass exposes raw coordinates without requiring inversion of the quantum circuit. We audit this mechanism using two tabular datasets, four architectures, and metrics conditioned on feature visibility. Iterative gradient matching gives median full-record PSNR of 73-96 dB for residual and input-only heads. Quantum-only heads score 8-11 dB on the full record but 54-96 dB on the six input coordinates they actually encode. These are reconstruction results for the tested six-input, six-observable circuits, not a general statement about quantum encodings. A loss-threshold membership attack remains near chance. The contribution is a visibility-conditioned privacy audit: omitted coordinates must not be credited as protection supplied by quantum processing, and near-exact PSNR differences must not be interpreted as meaningful privacy rankings. Our findings concern individual gradients and do not establish leakage under aggregation or multiple local training steps. Tema degli esploratori: fiducia digitale e crittografia (retro-catalogo arXiv: annunci del 2026-10-01 in cs.CR). Categorie arXiv: Quantum Physics (quant-ph) ; Cryptography and Security (cs.CR). Pagina del lavoro: https://arxiv.org/abs/2609.38720 — 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 2026-10-07T13:16:25.250Z dalla pagina https://arxiv.org/catchup/cs.CR/2026-10-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).