Formula

Probability Normalization over a Candidate Set

The conditional probability of a specific token, given a context, can be determined by normalizing its score against the scores of other tokens. This is achieved by dividing the score of the target token yiy_i by the sum of the scores for all tokens yjy_j within a defined candidate set ViV_i. This method ensures the resulting probabilities for all tokens in the set sum to 1. The general formula is: Pr(yicontext)=score(yi)yjViscore(yj)\text{Pr}(y_i | \text{context}) = \frac{\text{score}(y_i)}{\sum_{y_j \in V_i} \text{score}(y_j)}

0

1

Updated 2025-10-08

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