Metraipolis › archive › dc_5611c8bf55d7

Tracing model-generated DNA with position-independent watermarking — arXiv:2610.04763

Source
arXiv
Original
https://arxiv.org/abs/2610.04763
License (as archived)
CC0-1.0
Archived
2026-10-06T05:30:01.642Z
SHA-256 of the archived text
3366749938870d403b7bb8e963fefeb232fb4f5fb2025c581f0a51c89b0fedef
Stable URL of this page
https://metraipolis.com/documento/dc_5611c8bf55d7

How to cite

Tracing model-generated DNA with position-independent watermarking — arXiv:2610.04763 — arXiv, https://arxiv.org/abs/2610.04763 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5611c8bf55d7 (Metraipolis archive dc_5611c8bf55d7, sha256 3366749938870d403b7bb8e963fefeb232fb4f5fb2025c581f0a51c89b0fedef)
BibTeX
@misc{dc5611c8bf55d7,
  title = {Tracing model-generated DNA with position-independent watermarking — arXiv:2610.04763},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.04763},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5611c8bf55d7; sha256 3366749938870d403b7bb8e963fefeb232fb4f5fb2025c581f0a51c89b0fedef},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.04763. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Genomic language models can write synthetic DNA that carries no record of its origin. A generation-time watermark could provide a provenance signal, if a verifier can detect it using only the DNA sequence, a key, and published detector settings, without knowing where the generated region starts, which strand it lies on, or how it was divided into six-base tokens. We embed the SynthID tournament watermark in two genomic language models, Carbon and GENERator-v2. The verifier searches both strands, every start position, and four window lengths, and corrects its decision for the whole search. In a detection cohort of 1,544 held-out prompts, it found all 3,088 marked sequences per model, unedited and after one substituted, inserted, or deleted base. For ordinary sequences, the one-sided 95% upper confidence bound on the false-positive rate was at most 0.850%; one bound for the wrong-key control reached 1.015% (Carbon, after a deletion). In a development cohort, neither model showed a detectable change in likelihood or in predefined sequence measures. Under random edits placed without reference to the detector, detection stayed complete or nearly complete up to a 2% per-base edit rate. These results show statistical detection, not biological function, secret-key security, or robustness to an editor who sees the detector.

Tema degli esploratori: fiducia digitale e crittografia (elenco arXiv cs.CR, nuovi lavori del giorno).
Categorie arXiv: Genomics (q-bio.GN) ; Cryptography and Security (cs.CR).
Pagina del lavoro: https://arxiv.org/abs/2610.04763 — 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 2026-10-06T05:26:35.384Z dalla pagina https://arxiv.org/list/cs.CR/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).