Metraipolis › archive › dc_ff3daa2d9172

TextSeal2: Distortion-Free LLM Watermarking with Copula Mixing and Memoryless Races — arXiv:2610.40613

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.40613
License (as archived)
CC0-1.0
Archived
2026-10-31T06:15:00.351Z
SHA-256 of the archived text
7d6f3d9de42f6a0fa5897fc7289d901f5687cd263cf4086ff1343783ba2634eb
Stable URL of this page
https://metraipolis.com/documento/dc_ff3daa2d9172

How to cite

TextSeal2: Distortion-Free LLM Watermarking with Copula Mixing and Memoryless Races — arXiv:2610.40613 — arXiv (en), https://arxiv.org/abs/2610.40613 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ff3daa2d9172 (Metraipolis archive dc_ff3daa2d9172, sha256 7d6f3d9de42f6a0fa5897fc7289d901f5687cd263cf4086ff1343783ba2634eb)
BibTeX
@misc{dcff3daa2d9172,
  title = {TextSeal2: Distortion-Free LLM Watermarking with Copula Mixing and Memoryless Races — arXiv:2610.40613},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.40613},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ff3daa2d9172; sha256 7d6f3d9de42f6a0fa5897fc7289d901f5687cd263cf4086ff1343783ba2634eb},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.40613. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We present TextSeal2, a distortion-free watermark for large language models that keeps its signal under speculative decoding: tokens that replace rejected drafts remain watermarked with the same keys, and the acceptance rate is unchanged. It also preserves output diversity and adds no measurable inference latency. Its predecessor, TextSeal, restores diversity in the Gumbel-max scheme by routing each token to one of two keys via a coin flip. This is suboptimal because the detector averages the scores from both keys without knowing which key was used. TextSeal2 instead merges two or more keys into a single virtual key and couples it to fresh noise through AutoMark's Gaussian copula. Any number of keys, including publicly shared keys, can be merged this way without diluting detection. Under speculative decoding, TextSeal2 verifies keyed drafts with one more exponential race, run on the leftovers of the draft races. By memorylessness, this race reproduces block verification, so the output follows the target model exactly. We evaluate TextSeal2 on three LLMs. At equal output token diversity, TextSeal2 detects 8 to 22 percentage points more 100-token answers than TextSeal at a false-positive rate of $10^{-3}$, and is comparable to textGrain while requiring no post-logits optimization. Under speculative decoding, it detects 18 and 22 percentage points more 200-token answers than TextSeal and textGrain, respectively, at the same acceptance rate.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI); Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.40613 — 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 2026-10-31T06:07:08.782Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).