Abstract
Art therapy plays a vital role in emotional healing, with narrative creation serving as the primary vehicle for emotional expression. Given the dynamic nature of emotions during healing, narratives with finely controlled emotional fluctuations enable individuals to safely project inner conflicts and achieve emotional catharsis. Recently, the rapid development of Large Language Models (LLMs) has provided a new pathway for automated narrative generation technology to support such artistic designs. However, existing methods can produce fluent texts but struggle to generate narratives that adhere to specified affective trajectories, failing to meet the demands of emotion-oriented psychological healing.
To address these issues, this paper proposes EC-Script, an LLM agent-based framework that enables hierarchical control of the affective trajectory in narrative generation for emotional healing. EC-Script establishes overall narrative direction through Emotion-Trajectory Planning, propels scene-level plot development with Character-Driven Scene Generation, and regulates local emotional changes of characters via Emotion-Controlled Script Writing. Ultimately, it outputs scene-by-scene script content that remains highly consistent with the preset affective trajectory. Experimental results demonstrate that EC-Script significantly outperforms baseline methods in affective trajectory adherence, exhibiting excellent and reliable emotional controllability, thereby providing effective technical support for AI-assisted emotional healing scenarios.
Blogger's Review: The EC-Script framework proposed in this paper offers an innovative solution for emotional therapy, significantly enhancing the quality and emotional consistency of narrative generation through a hierarchical emotional control mechanism. This method not only holds practical application value but also points the way for future emotional AI research, showcasing the potential of LLMs in the mental health domain that deserves further exploration and implementation.