Metraipolis › archive › dc_2a243c6e89af
Watermarkable Multi-Draft Speculative Sampling via Poisson Processes — arXiv:2609.21858
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.21858
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-10T01:43:48.970Z
- SHA-256 of the archived text
2ba1990ba16b9952e9353f1b2c931fc474caeacf1fcb83d32a8e085c05eb4f17- Stable URL of this page
https://metraipolis.com/documento/dc_2a243c6e89af
How to cite
Watermarkable Multi-Draft Speculative Sampling via Poisson Processes — arXiv:2609.21858 — arXiv (en), https://arxiv.org/abs/2609.21858 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_2a243c6e89af (Metraipolis archive dc_2a243c6e89af, sha256 2ba1990ba16b9952e9353f1b2c931fc474caeacf1fcb83d32a8e085c05eb4f17)
BibTeX
@misc{dc2a243c6e89af,
title = {Watermarkable Multi-Draft Speculative Sampling via Poisson Processes — arXiv:2609.21858},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.21858},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_2a243c6e89af; sha256 2ba1990ba16b9952e9353f1b2c931fc474caeacf1fcb83d32a8e085c05eb4f17},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.21858. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large language models (LLMs) have achieved state-of-the-art performance across a wide range of tasks, motivating two important aspects of deployment: inference efficiency and output provenance, which can be tackled by speculative sampling and watermarking, respectively. However, recent works have shown that combining these two goals is highly nontrivial and can be potentially impossible. In this work, we develop a novel multi-draft speculative sampling algorithm based on Poisson processes that improves the frontier of this fundamental trade-off. The proposed algorithm has strong sampling efficiency on its own and, more interestingly, is naturally watermarkable: we can embed an unbiased watermark without degrading speculative acceptance. Moreover, our algorithm is based on an exact list-coupling-without-communication scheme, which yields a drafter invariance property that benefits both sampling and watermarking. It is the first multi-draft, drafter-invariant speculative sampling scheme that maintains both watermark strength and sampling efficiency, and we experimentally verify its strong performance in both aspects. Tema degli esploratori: fiducia digitale e crittografia (retro-catalogo arXiv: annunci del 2026-09-21 in cs.CR). Categorie arXiv: Cryptography and Security (cs.CR) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.21858 — 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 2026-10-10T01:13:42.033Z dalla pagina https://arxiv.org/catchup/cs.CR/2026-09-21?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).