Joint Detection and Location of English Puns

Code for the NAACL-19 paper: Joint Detection and Location of English Puns. This paper proposes to jointly address pun detection and location tasks by a sequence labeling approach with a newly designed tagging scheme.

Requirements

Python 3.6 Pytorch 0.4

Word embedding

Download the pretrained word embeddings glove.6B.100d.txt. Put the file under the folder embeddings/.

Reproducing the experimental results

To reproduce the results, simply do the following command:

bash run.sh

Cite

@InProceedings{zou-19-joint,
  author    = {Zou, Yanyan and Lu, Wei},
  title     = {Joint Detection and Location of English Puns},
  booktitle = {Proceedings of NAACL},
  year={2019}
}

## Contact

Yanyan Zou and Wei Lu, Singapore University of Technology and Design

Please feel free to drop an email at yanyan_zou@mymail.sutd.edu.sg for questions.

This implementation is inspired by the work