dogwhistle

Baseline code for NAACL 2021 paper "Blow the Dog Whistle: A Chinese Dataset for Cant Understanding with Common Sense and World Knowledge".

Download Data

Please read the terms here before you download the data. The dataset is distributed under the CC-BY-NC 3.0 license. The public data include train, dev, test splits.

Download the data here.

Leaderboard

https://competitions.codalab.org/competitions/30451#results

Submit to Leaderboard

Please refer to the description here.

Baseline

This repo has the baseline code we used in our experiments. It is based on a multi-choice MRC framework. However, we encourage you to try other types of models, e.g., two-tower model.

Citation

@inproceedings{dogwhistle,
author    = {Canwen Xu and
             Wangchunshu Zhou and
             Tao Ge and
             Ke Xu and
             Julian McAuley and
             Furu Wei},
title     = {Blow the Dog Whistle: A Chinese Dataset for Cant Understanding with Common Sense and World Knowledge},
booktitle = {{NAACL}},
year      = {2021}
}