Activity (Process)

Generating N-Best Candidates in BoN Sampling

In Best-of-NN (BoN) sampling, the initial step involves generating a set of NN candidate outputs that aim to maximize the conditional probability Pr(yx)\Pr(\mathbf{y}|\mathbf{x}). The method for producing these candidates can vary, depending on the search algorithm used by the model. Common techniques include stochastic methods like sampling or more deterministic approaches such as beam search.

0

1

Updated 2026-05-03

Contributors are:

Who are from:

Tags

Ch.5 Inference - Foundations of Large Language Models

Foundations of Large Language Models

Foundations of Large Language Models Course

Computing Sciences

Ch.4 Alignment - Foundations of Large Language Models