Learn Before
Concept

Objective of Supervised Fine-Tuning

The primary goal of Supervised Fine-Tuning (SFT) is to adapt a model that already has pre-trained parameters, denoted as θ^\hat{\theta}. The adaptation process involves adjusting these parameters to maximize the conditional probability of generating the desired output sequence, yy, given a specific input sequence, xx. This objective, expressed as maximizing Pr(yx)Pr(y|x), aligns the model's behavior with the provided instruction-response examples.

0

1

Updated 2026-05-02

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