Metraipolis › archive › dc_0673bd7e40cd

Backward Compatibility in Zero-Bit Neural Audio Watermarking — arXiv:2610.22296

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.22296
License (as archived): CC0-1.0
Archived: 2026-10-16T10:25:24.492Z
SHA-256 of the archived text: 372b608ad6e185289c1f9211b9eef9722c2e37fb71e29eabbca6c95d22206432
Stable URL of this page: https://metraipolis.com/documento/dc_0673bd7e40cd

How to cite

Backward Compatibility in Zero-Bit Neural Audio Watermarking — arXiv:2610.22296 — arXiv (en), https://arxiv.org/abs/2610.22296 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0673bd7e40cd (Metraipolis archive dc_0673bd7e40cd, sha256 372b608ad6e185289c1f9211b9eef9722c2e37fb71e29eabbca6c95d22206432)

BibTeX

@misc{dc0673bd7e40cd,
  title = {Backward Compatibility in Zero-Bit Neural Audio Watermarking — arXiv:2610.22296},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.22296},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0673bd7e40cd; sha256 372b608ad6e185289c1f9211b9eef9722c2e37fb71e29eabbca6c95d22206432},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.22296. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Neural watermarking systems are typically trained as tightly coupled embedder-detector pairs and implicitly assume static deployment. In real-world deployments, embedders and detectors are frequently updated and selectively decoupled for security and versioning purposes, which can break detection in previously embedded content and, more critically, increase false-positive rates on embedders that a given detector is intentionally not intended to recognize. In this paper the concept of backward compatibility in zero-bit audio watermarking is formalized. A training framework is proposed that pairs a new detector with a sequence of frozen legacy embedders (its legacy chain), using a performance-weighted sampling strategy so that the detector retains accuracy across all earlier versions. This coupling, while necessary for backward compatibility, is shown to induce an embedder-collapse effect that catastrophically degrades the true negative rate (TNR) against out-of-chain embedders. To counteract this, a Angular Diversity Loss (ADL) is introduced that explicitly forces each new embedder to occupy a distinct region of the output space. The combined approach restores near-perfect TNR with no measurable cost to detection accuracy or perceptual audio quality.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Sound (cs.SD) ; Machine Learning (cs.LG); Multimedia (cs.MM).
Pagina del lavoro: https://arxiv.org/abs/2610.22296 — 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 2026-10-16T08:48:02.948Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).