Metraipolis › archive › dc_c092207771a8
LAND: Learning an Adaptive Number of Draws for Bayesian Best-of-N Selection — arXiv:2611.03388
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.03388
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T10:33:47.546Z
- SHA-256 of the archived text
725a481304031087a691c250a1fefe1298999676ec8d92251f8ff0e0f4706583- Stable URL of this page
https://metraipolis.com/documento/dc_c092207771a8
How to cite
LAND: Learning an Adaptive Number of Draws for Bayesian Best-of-N Selection — arXiv:2611.03388 — arXiv (en), https://arxiv.org/abs/2611.03388 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c092207771a8 (Metraipolis archive dc_c092207771a8, sha256 725a481304031087a691c250a1fefe1298999676ec8d92251f8ff0e0f4706583)
BibTeX
@misc{dcc092207771a8,
title = {LAND: Learning an Adaptive Number of Draws for Bayesian Best-of-N Selection — arXiv:2611.03388},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.03388},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c092207771a8; sha256 725a481304031087a691c250a1fefe1298999676ec8d92251f8ff0e0f4706583},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.03388. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Best-of-N decoding improves generation quality by sampling multiple candidates and selecting the best, but it allocates equal compute to sources regardless of how much they benefit from additional sampling: the expected benefit is small for sources with low-variance score distributions and potentially much larger for those with high-variance distributions. To address this shortcoming, we propose LAND (Learning an Adaptive Number of Draws), which dynamically allocates samples according to their posterior-predictive marginal value. LAND learns a discrete empirical mixture of source-conditioned score distributions from calibration data, updates its belief about a source as completion scores are observed, and stops sampling when the expected improvement of another draw falls below a shared compute price. Across machine-translation experiments with multiple models and language pairs, LAND achieves better MetricX scores than fixed best-of-N at matched average compute. Conversely, at matched translation quality, LAND requires approximately 20-40% fewer candidate generations. Adaptive allocation therefore provides a simple way to retain the benefits of best-of-N while substantially improving its compute efficiency. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2611.03388 — 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 2026-11-03T10:29:48.637Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).