NAR-PMI

This repository contains the code for Findings of ACL2021 paper: A non-autoregressive edit-based approach to controllable text simplification.

Running the models

Scripts to train the models reported in Table 3 can be found in scripts/run_experiments.sh. The PMI constraints can be generated using notebooks/PMI.ipynb.

Cite the work

If you make use of the code, models, or algorithm, please cite our paper:

@inproceedings{agrawal-etal-2021-non,
    title = "A Non-Autoregressive Edit-Based Approach to Controllable Text Simplification",
    author = "Agrawal, Sweta  and
      Xu, Weijia  and
      Carpuat, Marine",
    booktitle = "Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021",
    month = aug,
    year = "2021",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2021.findings-acl.330",
    doi = "10.18653/v1/2021.findings-acl.330",
    pages = "3757--3769",
}