README.md

DetectGPT: Zero-Shot Machine-Generated Text Detection using Probability Curvature

Final Project COS484

We experiment changing two hyperparameters of the mode: the masked percentage and the temperature of the mask filling model.

Description of Relevant Files

• detect-gpt-writeup.pdf: This is the writeup of our hyperparameter tuning experiment.
• run.py: This contains the script that runs testing of DetectGPT.
• paper_scripts/: This directory contains the calls to run.py for each experiment.

Official implementation of the experiments in the DetectGPT paper.

An interactive demo of DetectGPT can be found here.

Instructions

First, install the Python dependencies:

python3 -m venv env
source env/bin/activate
pip install -r requirements.txt

Second, run any of the scripts (or just individual commands) in paper_scripts/.

If you'd like to run the WritingPrompts experiments, you'll need to download the WritingPrompts data from here. Save the data into a directory data/writingPrompts.

Note: Intermediate results are saved in tmp_results/. If your experiment completes successfully, the results will be moved into the results/ directory.

Citing the paper

If our work is useful for your own, you can cite us with the following BibTex entry:

@misc{mitchell2023detectgpt,
    url = {https://arxiv.org/abs/2301.11305},
    author = {Mitchell, Eric and Lee, Yoonho and Khazatsky, Alexander and Manning, Christopher D. and Finn, Chelsea},
    title = {DetectGPT: Zero-Shot Machine-Generated Text Detection using Probability Curvature},
    publisher = {arXiv},
    year = {2023},
}