Formula

Predictive Inference in Large Language Models

In the context of a Large Language Model (LLM), predictive inference involves selecting an output y^\hat{y} by optimizing the model's probability distribution. This is formally expressed by combining the general prediction formula, which starts as y^=arg...\hat{y} = \text{arg...}, with the specific probability function of the model, denoted as LLM Prθs()\text{LLM Pr}_\theta^s(\cdot). In this notation, θ\theta represents the model's parameters, and the superscript s may refer to a specific scoring method.

Image 0

0

1

Updated 2025-10-08

Contributors are:

Who are from:

Tags

Ch.4 Alignment - Foundations of Large Language Models

Foundations of Large Language Models

Foundations of Large Language Models Course

Computing Sciences