https://github.com/yzhangcs/ctc-copy

[EMNLP'23] Code for "Non-autoregressive Text Editing with Copy-aware Latent Alignments".
https://github.com/yzhangcs/ctc-copy

ctc non-autoregressive text-editing

Last synced: 5 months ago
JSON representation

[EMNLP'23] Code for "Non-autoregressive Text Editing with Copy-aware Latent Alignments".

• Host: GitHub
• URL: https://github.com/yzhangcs/ctc-copy
• Owner: yzhangcs
• Created: 2023-10-08T06:46:04.000Z (over 2 years ago)
• Default Branch: main
• Last Pushed: 2023-10-17T17:46:47.000Z (over 2 years ago)
• Last Synced: 2023-10-17T18:50:51.179Z (over 2 years ago)
• Topics: ctc, non-autoregressive, text-editing
• Language: Python
• Homepage: https://arxiv.org/abs/2310.07821
• Size: 50.8 KB
• Stars: 9
• Watchers: 1
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          


# Non-autoregressive Text Editing with Copy-aware Latent Alignments



  Yu Zhang1* 

  Yue Zhang1* 

  Leyang Cui2 

  Guohong Fu1 





1Soochow University, Suzhou, China



2Tencent AI Lab






[![conf](https://img.shields.io/badge/EMNLP%202023-orange?style=flat-square)](https://yzhang.site/assets/pubs/emnlp/2023/ctc.pdf)

[![arxiv](https://img.shields.io/badge/arXiv-2310.07821-b31b1b.svg?style=flat-square)](https://arxiv.org/abs/2310.07821)

[![citation](https://img.shields.io/badge/dynamic/json?label=citation&query=citationCount&url=https%3A%2F%2Fapi.semanticscholar.org%2Fgraph%2Fv1%2Fpaper%2F116277fd27c97d50bba2d8023d3c590c1ea8187b%3Ffields%3DcitationCount&style=flat-square)](https://www.semanticscholar.org/paper/Non-autoregressive-Text-Editing-with-Copy-aware-Zhang-Zhang/116277fd27c97d50bba2d8023d3c590c1ea8187b)

![python](https://img.shields.io/badge/python-%3E%3D%203.7-pybadges.svg?logo=python&style=flat-square)









## Citation

If you are interested in our work, please cite

```bib

@inproceedings{zhang-etal-2023-ctc,

  title     = {Non-autoregressive Text Editing with Copy-aware Latent Alignments},

  author    = {Zhang, Yu  and

               Zhang, Yue  and

               Cui, Leyang  and

               Fu, Guohong},

  booktitle = {Proceedings of EMNLP},

  year      = {2023},

  address   = {Singapore}

}

```

## Setup

The following packages should be installed:

* [`PyTorch`](https://github.com/pytorch/pytorch): >= 2.0

* [`Transformers`](https://github.com/huggingface/transformers)

* [`Errant`](https://github.com/chrisjbryant/errant)

Clone this repo recursively:

```sh

git clone https://github.com/yzhangcs/ctc-copy.git --recursive

```

You can follow this [repo](https://github.com/HillZhang1999/SynGEC) to obtain the 3-stage train/dev/test data for training a English GEC model.

The multilingual datasets are available [here](https://github.com/google-research-datasets/clang8).

Before running, you are required to preprocess each sentence pair into the format of `SRC:\t[src]\nTGT:\t[tgt]\n`, where `src` and `tgt` are the source and target sentences, respectively. Each sentence pair is separated by a blank line.

See [`data/clang8.toy`](data/clang8.toy) for examples.

## Run

Try the following command to train a 3-stage English model,

```sh

bash train.sh

```

To make predictions & evaluations:

```sh

bash pred.sh

```

## Contact

If you have any questions, please feel free to [email](mailto:yzhang.cs@outlook.com) me.