Summary of Multiverse Of Greatness: Generating Story Branches with Llms, by Pittawat Taveekitworachai et al.
Multiverse of Greatness: Generating Story Branches with LLMs
by Pittawat Taveekitworachai, Chollakorn Nimpattanavong, Mustafa Can Gursesli, Antonio Lanata, Andrea Guazzini, Ruck Thawonmas
First submitted to arxiv on: 22 Nov 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed Dynamic Context Prompting/Programming (DCP/P) framework leverages large language models (LLMs) to generate graph-based content with a dynamic context window history. This approach differs from previous studies that relied on manual processes and lacked flexibility in generating coherent stories. The DCP/P framework is evaluated against a baseline method that provides only the initial story data, without context history. Results show that providing LLMs with a summary leads to subpar storytelling compared to incorporating context history. Additionally, qualitative analysis reveals biases in word choices and sentiment of generated content, consistent with previous studies. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper creates a new way for computers to generate stories by using large language models (LLMs). The model, called Dynamic Context Prompting/Programming (DCP/P), lets LLMs create stories based on what happened before. This is different from other methods that required humans to manually write down the story and didn’t let the computer learn and adapt as it went along. The paper compares DCP/P with a simpler method and shows that providing context helps the computer generate better stories. |
Keywords
» Artificial intelligence » Context window » Prompting
