cs.AINov 2, 2026

LAND: Learning an Adaptive Number of Draws for Bayesian Best-of-N Selection

Authors: Koen Oostermeijer, Teryn Jones

Organizations: Aleph Alpha Research

Abstract

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.

Figures & tables

Explore similar work

CardsList
  1. Latency-Aware Adaptive Best-of-NN

    Nov 5, 2026Alon Jacobson, Qian Xie, Peter I. FrazierInference-Time Compute AllocationCost-Aware Inference

  2. Revisiting the (Sub)Optimality of Best-of-N for Inference-Time Alignment

    Mar 5, 2026Ved Sriraman, Adam BlockBest-of-N SamplingInference-Time Language Model Alignment

  3. Test-Time Scaling in the Wild: Why Exploitation, Not Exploration, Is the Bottleneck

    Aug 19, 2026Davide Romano, Kanak Raj, Jerrod Parker +1Test-Time ScalingLanguage Model Decoding