This repo contains codes for the following paper:

Jiaao Chen, Dinghan Shen, Weizhu Chen, Diyi Yang: HiddenCut: Simple Data Augmentation for Natural Language Understanding with Better Generalization. ACL 2021

If you would like to refer to it, please cite the paper mentioned above.

• CUDA, cudnn
• Python 3.7
• PyTorch 1.4.0

1. Install Huggingface Transformers according to the instructions here: https://github.com/huggingface/transformers.

2. Download the datasets from the GLUE benchmark:

python download_glue_data.py --data_dir glue_data --tasks all

3. Fine-tune the RoBERTa model with the HiddenCut data augmentation strategies:

>>> chmod +x run_glue_hidden_cutoff.sh
>>> ./run_glue_hidden_cutoff.sh

This repository is built upon https://github.com/dinghanshen/Cutoff