Loose Ends: A Mixed-Initiative Creative Interface for Playful Storytelling

Abstract

We present Loose Ends, a mixed-initiative co-creative storytelling play experience in which a human player and an AI system work together to compose a story. Loose Ends specifically aims to provide computational support for managing multiple parallel plot threads and bringing these threads to satisfying conclusions—something that has proven difficult in past attempts to facilitate playful mixed-initiative storytelling. We describe the overall human-AI interaction loop in Loose Ends, including the implementation of the rules-based AI system that enables this interaction loop; discuss four examples of desirable mixed-initiative interactions that are possible in Loose Ends, but not in similar systems; and present results from a preliminary expert evaluation of Loose Ends. Altogether, we find that Loose Ends shows promise for creating a sense of coauthorship in the player while also mitigating the directionlessness reported by players of earlier systems.

Authors' Values and Attitudes Towards AI-bridged Scalable Personalization of Creative Language Arts

“…While generation of CLA has been explored with various technical backbones, including template-based approaches [19,39], symbolic planning [56,65,81,87,98], casebased reasoning [38,83,85,91,95], or character simulation [15,61], LLM technologies powered by transformer architecture [96] have brought in a large leap in flexibility and accuracy of text generation, as these models could be "prompted" to serve arbitrary natural language tasks [12,74,94]. With these LLM capabilities, many writing tools have been introduced, and one type of tool is those that suggest text phrases to the user's writing, which is often called human-AI co-writing [2,36,53,66,103]. Researchers studied how these LLM suggestions can change people's writing and found that generated texts could spark new ideas [14,18,36], lower grammatical errors [58], and increase vocabulary diversity [58], while introducing cognitive challenge of integrating generated texts into the user's writing [14,90].…”

The Right Variety: Improving Expressive Range Analysis with Metric Selection Methods

“…Cook et al developed the Danesh system for conducting automatic ERA within Unity, as well as for identifying the trends in ERA across alternative parameterisations of the same generator using what they termed Randomised ERA, or RERA [5]. Kremenski et al introduced 'Expressive Range Coverage Analysis' to visualise trends in how designers interacted with a generative system over time [15]. We have also explored the use of alternative techniques such as dimensionality reduction and Convolutional Neural Network embeddings to produce ERA style generative space visualisations without the need for defining or recalculating the type of diversity we are interested in [25,26].…”

“…The alternative approach for metric selection is to not select at all, and instead visualise the plots for all metric pairs that might be of interest. This longer list of metrics can again be sourced inclusively from the prior literature on similar game domains [2,11], or a list can be designed by a game domain expert [15]. This list of n metrics can then be visualised using ERA of all possible pairs, commonly using Summerville's corner plot concept [22].…”

“…These 2D plots can then be used to understand the character and diversity present in individual generative spaces, but also those from alternative generators compare to each other. ERA has most commonly been applied to work within its original domain of PCG for game levels, but it has also found a place in alternative areas of PCG research such as visualising the output of poetry generators [15], tabletop game outcomes [10] and emergent narrative systems [16].…”

AIsop: Exploring Immersive VR Storytelling Leveraging Generative AI

“…Notable early examples of VR storytelling, such as "Son of Jaguar" and "Dear Angelica" [6] exemplify how narrative elements are effectively interwoven with visual representations in virtual environments, demonstrating the unique capabilities of VR in enhancing the storytelling experience. AI's role in storytelling extends across multiple facets, ranging from assisting narrators in maintaining coherence in the main plot and its subplots [8], to predicting characters' emotional arcs [1]. In interactive mediums like computer games, AI has been instrumental in developing behaviors for Non-Playable Characters (NPCs) [4], as well as in generating dynamic NPC dialogues [3].…”