The data for our EMNLP 2022 paper arXivEdits: Understanding the Human Revision Process in Scientific Writing is provided at this repo.

The name of each field should be self-explainable. If you have any questions, please reach me at chaojiang06@gmail.com.

The code for extracting plain text from latex source code was written by awesome Sam Stevens when he was an undergraduate student. The raw code can be found here.

We add a pipeline code. Input paragraph pairs; it will do sentence alignment, edit extraction, and intention classification for you. Your one-stop solution for revision analysis. Check it out!

We upload all fine-tuned T5 intention classification models to the huggingface hub. The code is in the code folder

We upload all fine-tuned BERT checkpoints to the huggingface hub and provide a sample code to use them.

We add license information every version of all papers. For example:

    train['1608.00087']['license'] = {'1': 'http://arxiv.org/licenses/nonexclusive-distrib/1.0/', '2': 'http://arxiv.org/licenses/nonexclusive-distrib/1.0/'}

In total, we find the following licenses:

    http://arxiv.org/licenses/assumed-1991-2003/
    http://arxiv.org/licenses/nonexclusive-distrib/1.0/
    http://creativecommons.org/licenses/by-nc-sa/4.0/
    http://creativecommons.org/licenses/by-sa/4.0/
    http://creativecommons.org/licenses/by/3.0/
    http://creativecommons.org/licenses/by/4.0/
    http://creativecommons.org/licenses/publicdomain/
    http://creativecommons.org/publicdomain/zero/1.0/

We also add the source arxiv-id for each sentence pair in the edits sub-dataset.

Thanks for the suggestions from Qian Ruan from the UKP Lab!

If you find our paper or dataset useful, please considering cite the following paper.

@article{jiang-etal-2022-arXivEdits,
  title={arXivEdits: Understanding the Human Revision Process in Scientific Writing},
  author={Jiang, Chao and Xu, Wei and Stevens, Samuel},
  journal={In Proceedings of EMNLP 2022},
  year={2022}
}