TransESC

The official implementation for the Findings of ACL 2023 paper TransESC: Smoothing Emotional Support Conversation via Turn-Level State Transition.

Requirements

• Python 3.7
• PyTorch 1.8.2
• Transformers 4.12.3
• CUDA 11.1

Preparation

Download the processed dataset, annotated emotion and keywords, and generated commonsense knowledge from https://drive.google.com/drive/folders/1rNpa5vDuB7KssQuVBAhkKtmAuqPa0Qgf?usp=sharing. Then place them in /dataset.

Training

python main.py

Evaluation and Generation

python main.py --test 

Citation

If you find our work useful for your research, please kindly cite our paper as follows:

@article{zhao2023transesc,
  title={TransESC: Smoothing Emotional Support Conversation via Turn-Level State Transition},
  author={Zhao, Weixiang and Zhao, Yanyan and Wang, Shilong and Qin, Bing},
  journal={arXiv preprint arXiv:2305.03296},
  year={2023}
}