Summary of Threads Of Subtlety: Detecting Machine-generated Texts Through Discourse Motifs, by Zae Myung Kim and Kwang Hee Lee and Preston Zhu and Vipul Raheja and Dongyeop Kang
Threads of Subtlety: Detecting Machine-Generated Texts Through Discourse Motifs
by Zae Myung Kim, Kwang Hee Lee, Preston Zhu, Vipul Raheja, Dongyeop Kang
First submitted to arxiv on: 16 Feb 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 paper explores the nuances of human-written texts by identifying distinctive linguistic properties that set them apart from machine-generated ones. The authors propose a novel methodology using hierarchical parse trees and recursive hypergraphs to uncover discourse patterns in both human-written and large language model (LLM)-generated texts. The results show that human-written texts exhibit more structural variability, reflecting the complexities of human writing. Furthermore, incorporating hierarchical discourse features improves the performance of binary classifiers in distinguishing between human-written and machine-generated texts. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is all about understanding what makes human-written texts unique. It’s like comparing apples and oranges – while both might look similar on the surface, they have different insides that make them special. The researchers developed a new way to study text patterns using special trees and graphs. They found that human-written texts are more varied and complex than those generated by machines. This is important because it helps us tell apart real human writing from machine-generated text. |
Keywords
» Artificial intelligence » Discourse » Large language model
