Concept

Token Selection from Probability Distribution

After a language model computes the probability distribution for the next token, Pr(·|x_0, ..., x_{i-1}), a specific token x_i must be chosen from this distribution. This selection process, also known as decoding or sampling, is a fundamental step in text generation.

0

1

Updated 2026-04-18

Contributors are:

Who are from:

Tags

Ch.2 Generative Models - Foundations of Large Language Models

Foundations of Large Language Models

Foundations of Large Language Models Course

Computing Sciences