Metraipolis › archive › dc_5d6326da2eb3
Testing Processor Contracts in Speculative Decoding Engines — arXiv:2611.04042
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.04042
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-04T12:41:24.527Z
- SHA-256 of the archived text
9a9cd25f57fa771aa37653e7fd8078e9cdac7e7c3a66147e32391adfcda7bb79- Stable URL of this page
https://metraipolis.com/documento/dc_5d6326da2eb3
How to cite
Testing Processor Contracts in Speculative Decoding Engines — arXiv:2611.04042 — arXiv (en), https://arxiv.org/abs/2611.04042 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5d6326da2eb3 (Metraipolis archive dc_5d6326da2eb3, sha256 9a9cd25f57fa771aa37653e7fd8078e9cdac7e7c3a66147e32391adfcda7bb79)
BibTeX
@misc{dc5d6326da2eb3,
title = {Testing Processor Contracts in Speculative Decoding Engines — arXiv:2611.04042},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.04042},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5d6326da2eb3; sha256 9a9cd25f57fa771aa37653e7fd8078e9cdac7e7c3a66147e32391adfcda7bb79},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.04042. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Speculative sampling can accelerate decoding without changing the target distribution. Its standard guarantee assumes that logits processors, which enforce constraints or modify token scores, receive the appropriate token prefix and state at each verification position. We formalize this processor-prefix contract and present BADC (branch-aware differential conformance testing). BADC combines enumerated references on tiny random models, checks of processor inputs and state, and simultaneous confidence bounds on output-event probabilities. Preregistered, source-informed experiments on selected Transformers and llama-cpp-python releases confirm two causes: processors read unverified drafts or advance state without rollback. For one pilot-selected prompt on a trained checkpoint, fresh samples give a 60.26-percentage-point lower bound on the difference in the probability of a predefined first-token event between ordinary and prompt-lookup decoding, with nominal simultaneous 95% coverage under within-arm i.i.d. sampling. Controlled comparisons establish no cause-recovery advantage of directed selection over random selection, while permutation testing detects each tested family at least as often as BADC. These findings connect processor-interface failures to observable decoding differences without establishing diagnostic superiority or workload prevalence. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Software Engineering (cs.SE) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2611.04042 — 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 2026-11-04T12:29:47.375Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).