Metraipolis › archive › dc_bebdded8d6e4

HaloMark: A Spectral Threshold for Embedding-Vector Watermarking under C2PA — arXiv:2608.08645

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.08645
License (as archived)
CC0-1.0
Archived
2026-10-19T08:58:16.362Z
SHA-256 of the archived text
996fcc8e12c47e1101f4f2b321a3f6c87dd342859c2f22dae1ed2a5711035279
Stable URL of this page
https://metraipolis.com/documento/dc_bebdded8d6e4

How to cite

HaloMark: A Spectral Threshold for Embedding-Vector Watermarking under C2PA — arXiv:2608.08645 — arXiv (en), https://arxiv.org/abs/2608.08645 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_bebdded8d6e4 (Metraipolis archive dc_bebdded8d6e4, sha256 996fcc8e12c47e1101f4f2b321a3f6c87dd342859c2f22dae1ed2a5711035279)
BibTeX
@misc{dcbebdded8d6e4,
  title = {HaloMark: A Spectral Threshold for Embedding-Vector Watermarking under C2PA — arXiv:2608.08645},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.08645},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_bebdded8d6e4; sha256 996fcc8e12c47e1101f4f2b321a3f6c87dd342859c2f22dae1ed2a5711035279},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.08645. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Foundation-model embeddings are now a primary data asset, but the content-provenance machinery built for images and audio does not transfer to them. C2PA binds to an asset with a stable bit-level or perceptual identity; embeddings have neither, since quantisation, projection, fine-tuning, and windowed averaging reshape them in normal use and break any fixed hash. We present HaloMark, a watermark for embedding vectors cryptographically bound to a C2PA manifest. It composes four standard primitives -- a block-diagonal orthogonal rotation, public whitening, an input-dependent LSH commitment, and a per-vector nonce -- around one protocol change: the producer signs the LSH commitment c into the C2PA sidecar, and the verifier reads c from the manifest instead of recomputing it. Recomputing is fragile under whitening, which flips the commitment bucket on 62% of inputs at cos = 0.96; reading the signed c reduces the verifier's score to T = T_null + beta(A)*epsilon, so security turns on a single scalar beta, which we bound rigorously for linear and non-adaptive attackers and characterise empirically for the adaptive case. We evaluate against an adversary holding polynomially many clean/watermarked pairs under one key with full sidecar visibility, across eight baselines and ten adaptive attackers including denoising-autoencoder removal. The eleven encoders separate at an empirical threshold eff_rank(Sigma)/d ~= 0.19: above it, detection AUROC stays at 0.98 or higher across every in-budget attack on the three encoders we sweep in full, and at 0.965 or higher under single-seed DAE removal on the rest; below it every variant we tested fails. Why the threshold is dimension-uniform is left open. Deployed as a Qdrant admission filter, the verifier runs at 284 us and 24 bytes of sidecar per vector, validated end-to-end against three C2PA reference-SDK bindings.

Tema degli esploratori: fiducia digitale e crittografia (retro-catalogo arXiv: annunci del 2026-08-11 in cs.CR).
Categorie arXiv: Cryptography and Security (cs.CR) ; Information Retrieval (cs.IR).
Pagina del lavoro: https://arxiv.org/abs/2608.08645 — 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 2026-10-19T08:54:24.948Z dalla pagina https://arxiv.org/catchup/cs.CR/2026-08-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).