Modeling Protagonist Emotions for Emotion-Aware Storytelling
This work addresses the challenge of emotion-aware storytelling for narrative generation systems, representing an incremental advancement in incorporating emotional modeling into automated story creation.
The paper tackled the problem of generating stories that follow specified emotional arcs for protagonists in neural storytelling, achieving significantly better adherence to desired emotion arcs compared to baseline methods without compromising story quality.
Emotions and their evolution play a central role in creating a captivating story. In this paper, we present the first study on modeling the emotional trajectory of the protagonist in neural storytelling. We design methods that generate stories that adhere to given story titles and desired emotion arcs for the protagonist. Our models include Emotion Supervision (EmoSup) and two Emotion-Reinforced (EmoRL) models. The EmoRL models use special rewards designed to regularize the story generation process through reinforcement learning. Our automatic and manual evaluations demonstrate that these models are significantly better at generating stories that follow the desired emotion arcs compared to baseline methods, without sacrificing story quality.