Segmenting Watermarked Texts From Language Models

Implementation of the methods described in "Segmenting Watermarked Texts From Language Models" by Xingchi Li, Guanxun Li, Xianyang Zhang.

Prerequisites

Python environments

• Cython==3.0.10
• datasets==2.19.1
• huggingface_hub==0.23.0
• nltk==3.8.1
• numpy==1.26.4
• sacremoses==0.0.53
• scipy==1.13.0
• sentencepiece==0.2.0
• tokenizers==0.19.1
• torch==2.3.0.post100
• torchaudio==2.3.0
• torchvision==0.18.0
• tqdm==4.66.4
• transformers==4.40.2

Set up environments

conda install pytorch torchvision torchaudio pytorch-cuda=12.1 -c pytorch -c nvidia
conda install conda-forge::transformers
conda install cython scipy nltk sentencepiece sacremoses

Instructions

All experiments are conducted using Slurm workload manager. Expected running time and memory usage are provided in the corresponding sbatch scripts.

Important

Please modify the paths, Slurm mail options and adjust the GPU resources in the sbatch scripts before running the experiments.

Caution

The Python SeedBS script is modified based on the R version. The output is not guaranteed to be the same.

# Setup pyx.
sbatch 1-setup.sh

# Text generation.
bash 2-textgen-helper.sh
sbatch 2-textgen.sh

# Rolling window watermark detection.
bash 3-detect-helper.sh
sbatch 3-detect.sh

# Change point analysis using R.
bash 4-seedbs-helper.sh
sbatch 4-seedbs.sh
# OR using Python.
bash 4.1-seedbs-helper.sh
sbatch 4.1-seedbs.sh

# Result analysis and ploting. (NOT)
Rscript 5-not.R

Tip

The implementation of NOT can be found in the 5-not.R script from line 348 to 371.

Citation

@inproceedings{
  li2024segmenting,
  title={Segmenting Watermarked Texts From Language Models},
  author={Xingchi Li and Guanxun Li and Xianyang Zhang},
  booktitle={The Thirty-eighth Annual Conference on Neural Information Processing Systems},
  year={2024},
  url={https://openreview.net/forum?id=FAuFpGeLmx}
}

Stargazers over time