README.md

metadata

license: apache-2.0
size_categories:
  - 10K<n<100K
task_categories:
  - text-classification
  - text-generation
language:
  - en
tags:
  - authorship obfuscation
  - authorship attribution
  - authorship classification
  - authorship style
  - style transfer
  - style
  - author

[StyleRemix] AuthorMix Dataset

Dataset Description

This contains the AuthorMix dataset, which is created for authorship obfuscation. It includes data from four distinct domains: presidential speeches, early-1900s fiction novels, scholarly articles, and diary-style blogs. Altogether, AuthorMix contains over 30k high-quality paragraphs from 14 authors.

This work was created in the paper: StyleRemix: Interpretable Authorship Obfuscation via Distillation and Perturbation of Style Elements (https://arxiv.org/abs/2408.15666v1), which should be cited if used in future work.

Please see our paper (https://arxiv.org/abs/2408.15666v1) or Github repository (https://github.com/jfisher52/StyleRemix) for more details!

Dataset Summary

For the presidential domain, we curate and clean speeches from George W. Bush, Barack Obama, and Donald Trump. For novel domain, we choose a collection of early 1900s fiction writers with strong writing styles: Ernest Hemingway, F. Scott Fitzgerald, and Virginia Woolf. We select these presidents/authors due to their diverse styles but similar eras to minimize content discrepancies. Lastly, we alter two existing datasets to match the formality of our new domains: the Extended-Brennan Greenstad, a collection of "scholarly" short (500-word) paragraphs gathered from Amazon Mechanical Turk (AMT), and the Blog Authorship corpus, a collection of blogs (diary-style entries) that were posted to blog.com.

Other Helpful Links

Please see this Huggingface Collection (https://huggingface.co/collections/hallisky/authorship-obfuscation-66564c1c1d59bb62eaaf954f) if you want to access any other artifacts from the StyleRemix paper.

Citing Our Work

Please cite our paper if you use this dataset or other resources:

@misc{fisher2024styleremixinterpretableauthorshipobfuscation,
      title={StyleRemix: Interpretable Authorship Obfuscation via Distillation and Perturbation of Style Elements}, 
      author={Jillian Fisher and Skyler Hallinan and Ximing Lu and Mitchell Gordon and Zaid Harchaoui and Yejin Choi},
      year={2024},
      eprint={2408.15666},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2408.15666}, 
}