Welcome to The Next Chapter: A Study of Large Language Models in Storytelling! This project collects data related to our research, which has been published in INLG2023.
The goal of this project is to explore the potential of large language models in storytelling, examplified by GPT-3. We have conducted extensive comparison and analysis, and you can find the details in our published paper.
The data collected for this study is organized into two main sections:
All the generated stories can be found within the GeneratedStories directory. Each text file contains a series of stories, with each line representing a unique pairing of "Condition" and "Story."
We have also included the results of the human evaluation in the HumanEvaluation directory. These results provide valuable insights into different quality aspects of the generated stories.
If you have any questions, feedback, feel free to reach out to Zhuohan Xie at [email protected]. We welcome any inquiries and look forward to hearing from you!