Summary of Hullmi: Human Vs Llm Identification with Explainability, by Prathamesh Dinesh Joshi et al.
HULLMI: Human vs LLM identification with explainability
by Prathamesh Dinesh Joshi, Sahil Pocker, Raj Abhijit Dandekar, Rajat Dandekar, Sreedath Panat
First submitted to arxiv on: 7 Sep 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: None
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 study explores the effectiveness of traditional machine learning models compared to modern NLP detectors for detecting human versus AI-generated text. The research implements a robust testing procedure on diverse datasets, including curated corpora and real-world samples, showing that traditional ML models perform similarly to modern NLP detectors like T5-Sentinel and RoBERTa-Sentinel. To gain insights into the detection process, the study employs explainable AI technique LIME, uncovering parts of the input contributing most to each model’s prediction. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research compares traditional machine learning models with modern NLP detectors for identifying human-written versus AI-generated text. The results show that traditional models are just as good at detecting human or AI text as newer NLP detectors. To better understand how these models work, the study uses a technique called LIME to see which parts of the input text are most important for each model’s prediction. |
Keywords
» Artificial intelligence » Machine learning » Nlp » T5
