Abstract:This paper assesses the potential for large language models (LLMs) to serve as assistive tools in the creative writing process, by means of a single, in-depth case study. In the course of the study, we develop interactive and multi-voice prompting strategies that interleave background descriptions (scene setting, plot elements), instructions that guide composition, samples of text in the target style, and critical discussion of the given samples. We qualitatively evaluate the results from a literary critical perspective, as well as from the standpoint of computational creativity (a sub-field of artificial intelligence). Our findings lend support to the view that the sophistication of the results that can be achieved with an LLM mirrors the sophistication of the prompting.

What problem does this paper attempt to address?

This paper attempts to explore the potential of large - language models (LLMs) in the creative writing process and evaluate the effectiveness of LLMs as an auxiliary tool through an in - depth case study. Specifically, the paper focuses on the following aspects: 1. **Interactive and polyphonic prompting strategies**: Researchers developed prompting strategies that combine background descriptions (scene settings, plot elements), guiding instructions, text samples of the target style, and critical discussions of given samples to explore how these strategies affect the quality of the text generated by LLMs. 2. **Qualitative evaluation from the perspective of literary criticism**: In addition to evaluation from the perspective of computational creativity, researchers also conducted a qualitative analysis of the text generated by LLMs from the perspective of literary criticism, without considering its "authorship". 3. **The influence of the temperature parameter**: Researchers adjusted the temperature parameter of LLMs (from 0.9 to 1.4) to observe the changes in the generated text under different settings, especially in terms of experimental style and innovation. 4. **Multi - role generation**: Researchers transformed the text generated by the dialogue into a script with alternating author and tutor voices, and then used this script to prompt the model to play the roles of both author and tutor simultaneously for self - criticism. Through these methods, the paper aims to demonstrate the potential of LLMs in creative writing and explore the possibility of human - machine cooperation rather than replacing human creativity. The study found that the responses of LLMs were unexpectedly complex, able to continuously improve the text quality in the iterative process, and the new words generated were also semantically coherent and context - relevant. In addition, in the multi - role mode, the model even spontaneously introduced completely new roles, although there was no such hint in the prompt. Overall, through specific experiments and analyses, this paper shows the application prospects of LLMs in the field of creative writing and emphasizes the importance of human - machine collaboration.

Evaluating Large Language Model Creativity from a Literary Perspective

On the Creativity of Large Language Models

Creativity Support in the Age of Large Language Models: An Empirical Study Involving Emerging Writers

Steering Large Language Models to Evaluate and Amplify Creativity

Assessing and Understanding Creativity in Large Language Models

Art or Artifice? Large Language Models and the False Promise of Creativity

Large Language Models show both individual and collective creativity comparable to humans

Characterising the Creative Process in Humans and Large Language Models

Divergent Creativity in Humans and Large Language Models

Evaluating Creative Short Story Generation in Humans and Large Language Models

Human Creativity in the Age of LLMs: Randomized Experiments on Divergent and Convergent Thinking

Exploring the Potential of Large Language Models in Artistic Creation: Collaboration and Reflection on Creative Programming

Pron vs Prompt: Can Large Language Models already Challenge a World-Class Fiction Author at Creative Text Writing?

The Language of Creativity: Evidence from Humans and Large Language Models

LARGE LANGUAGE MODEL-BASED ARTIFICIAL INTELLIGENCE IN THE LANGUAGE CLASSROOM: PRACTICAL IDEAS FOR TEACHING

Small Language Models can Outperform Humans in Short Creative Writing: A Study Comparing SLMs with Humans and LLMs

HoLLMwood: Unleashing the Creativity of Large Language Models in Screenwriting via Role Playing

Assessing the Creativity of LLMs in Proposing Novel Solutions to Mathematical Problems

LLM Discussion: Enhancing the Creativity of Large Language Models via Discussion Framework and Role-Play

A Large Language Model-based Computational Approach to Improve Identity-Related Write-Ups