Steering Narrative Agents through a Dynamic Cognitive Framework for Guided Emergent Storytelling

Nov 10, 2025·
C. Yang
,
M. Gross
Dr. Rafael Wampfler
Dr. Rafael Wampfler
Abstract
We present a dynamic cognitive framework for steering narrative agents in interactive digital entertainment. The framework enables agents to pursue coherent, goal-directed story arcs while remaining responsive to emergent player actions and choices. Using belief-desire-intention (BDI) representations combined with large language model generation, narrative agents dynamically update their internal goals, beliefs, and story intentions in response to player input, balancing narrative coherence with interactive agency. We evaluate the framework in a series of multi-turn interactive storytelling scenarios, demonstrating that agents guided by the framework maintain more coherent narratives and generate more satisfying story outcomes compared to unguided baselines.
Type
Publication
In Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE), Edmonton, Canada
publications
Narrative AI Cognitive Architectures Conversational Agents Interactive Storytelling Large Language Models
Dr. Rafael Wampfler
Authors
Dr. Rafael Wampfler
Senior Researcher & Lecturer

I am a Senior Researcher & Lecturer at the Computer Graphics Laboratory of ETH Zurich, and a Research Consultant at Disney Research. I am leading the Digital Character AI projects at CGL. My research interests include conversational digital characters, affective computing, human-computer interaction, and applied machine learning.

My vision is to create intelligent digital humans that can naturally communicate, understand, and support people across domains such as education and mental health. My research focuses on multimodal artificial intelligence for interactive digital humans, developing models that combine large language models, affective computing, and data-driven animation to create embodied conversational agents endowed with autonomous agency, consistent values, and beliefs.

My work bridges machine learning, human–computer interaction, and computer graphics to enable AI systems such as Digital Einstein and interactive patient avatars for psychotherapy training and health education.