The field of interactive narrative systems is moving towards a more formal and structured approach, with a focus on developing frameworks and methodologies for analyzing and evaluating these systems. This shift is driven by the need for a more coherent and collaborative research community, and is being facilitated by advances in areas such as large language models and multimodal processing. One of the key challenges in this field is the ability to effectively capture and represent complex narrative structures, and to develop systems that can reason about and generate coherent stories. Recent work has highlighted the potential of large language models to augment and reimagine existing storyline visualization techniques, and to provide new insights into familiar stories. However, current models still lack genuine story-level intelligence, and struggle with non-linear narratives, character consistency, and causal inference across extended sequences. Noteworthy papers in this area include:

• A paper that introduces a formal representation framework for interactive narrative systems, which provides a consistent vocabulary and modeling structure for analyzing and comparing these systems.
• A paper that presents a novel evaluation framework for assessing the narrative understanding of vision language models, which reveals significant limitations in current models' ability to reason about temporal causality and cross-panel cohesion.