Total: 1
The use of large language models (LLMs) in human-centered creative tasks — such as journalism, scientific writing, and storytelling — has showcased their potential for content generation but highlighted a critical gap: planning. Planning, used here to describe the “actions” humans perform before (and during) the writing process, is a fundamental process in many creative domains. This tutorial explores how planning has been learned and deployed in creative workflows, unifying three scenarios: Full Data Regimens (when observational data for actions and the resulting text exist), Partial (when text exists but actions can be inferred) and Low (when neither exist). The tutorial discusses forward and backward learning approaches for planning in LLMs, evaluation metrics tailored to latent plans, and practical applications in computational journalism, web agents, and other creative domains. By bridging theoretical concepts and practical demonstrations, this tutorial aims to inspire new research directions in leveraging LLMs for creative and goal-oriented planning tasks.