Activity (Process)

Data Collection for Reward Modeling in RLHF

The initial step in training a reward model is to gather a dataset of human feedback. This process starts by using a Large Language Model to generate several different candidate outputs for a given input prompt, x. These outputs are then evaluated by humans, and their feedback can be collected using a variety of methods.

0

1

Updated 2025-10-10

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

Ch.4 Alignment - Foundations of Large Language Models