ImprovMate: Multimodal AI Assistant for Improv Actor Training

2025-06-29

AI Summary

Improvisational actors often experience excessive cognitive load when simultaneously maintaining narrative coherence and character continuity; existing AI-assisted approaches rely heavily on manual intervention and fail to effectively leverage large language models (LLMs). Method: This study introduces the first LLM-integrated framework for improvisational theater training—a modular GPTs system designed for professional practice. It employs dynamic exercise mechanisms and structured reference tables to calibrate generative randomness, and incorporates scenario-specific modules such as abrupt plot termination and reactive thinking training to enable automated, low-intrusion narrative prompting. Results: Empirical evaluation demonstrates that the system significantly reduces memory load while enhancing creative focus—without disrupting traditional pedagogical logic—and receives strong acceptance from practitioners regarding its novelty and pedagogical efficacy. The core contribution is a novel LLM-driven paradigm for improvisational training that balances generative controllability with artistic appropriateness.

Technology Category

• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Machine Learning: Large Multimodal Models (LMMs)
• Humans and AI: Game Design — Procedural Content Generation & Storytelling

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

Improvisation training for actors presents unique challenges, particularly in maintaining narrative coherence and managing cognitive load during performances. Previous research on AI in improvisation performance often predates advances in large language models (LLMs) and relies on human intervention. We introduce ImprovMate, which leverages LLMs as GPTs to automate the generation of narrative stimuli and cues, allowing actors to focus on creativity without keeping track of plot or character continuity. Based on insights from professional improvisers, ImprovMate incorporates exercises that mimic live training, such as abrupt story resolution and reactive thinking exercises, while maintaining coherence via reference tables. By balancing randomness and structured guidance, ImprovMate provides a groundbreaking tool for improv training. Our pilot study revealed that actors might embrace AI techniques if the latter mirrors traditional practices, and appreciate the fresh twist introduced by our approach with the AI-generated cues.

Problem

• Automating narrative stimuli generation for improv actors
• Reducing cognitive load during improvisation performances
• Balancing randomness and structured guidance in training

Innovation

• Leverages LLMs to automate narrative stimuli
• Incorporates exercises mimicking live training
• Balances randomness with structured guidance