Case Study

Designing a Data Collection Pipeline for a Creative Writing Assistant

You are an ML engineer at a startup building a creative writing assistant. To align the model with user preferences for 'helpful and imaginative' suggestions, you need to create a dataset for training a reward model. Describe a concrete, step-by-step process for this data collection phase. Your description should specify (1) what the language model will generate for a given input prompt (e.g., a story idea), and (2) what specific task the human labelers will perform with the generated content.

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

Application in Bloom's Taxonomy

Cognitive Psychology

Psychology

Social Science

Empirical Science

Science