Learn Before
Activity (Process)

Process of Beam Search

At the first time step, beam search selects the kk tokens with the highest predicted probabilities from the vocabulary to serve as the initial kk candidate sequences. At each subsequent time step, the algorithm considers the kk candidate sequences from the previous step and extends them with all possible tokens, creating k\left|\mathcal{Y} ight| possible combinations. From these combinations, it evaluates and selects the kk extended sequences with the highest overall predicted probabilities to carry forward to the next step.

0

1

Updated 2026-05-14

Contributors are:

Who are from:

Tags

D2L

Dive into Deep Learning @ D2L