A2Net_ECPE

Joint Alignment of Multi-Task Feature and Label Spaces for Emotion Cause Pair Extraction

This repository contains the code of the official implementation for the paper: Joint Alignment of Multi-Task Feature and Label Spaces for Emotion Cause Pair Extraction. The paper has been accepted to appear at Coling 2022.

Some code is based on Rank-Emotion-Cause, and Partition Filter Network.

If you use our codes or your research is related to our paper, please kindly cite our paper:

@inproceedings{chen-etal-2022-joint,
    title = "Joint Alignment of Multi-Task Feature and Label Spaces for Emotion Cause Pair Extraction",
    author = "Chen, Shunjie  and
      Shi, Xiaochuan  and
      Li, Jingye  and
      Wu, Shengqiong  and
      Fei, Hao  and
      Li, Fei  and
      Ji, Donghong",
    booktitle = "Proceedings of the 29th International Conference on Computational Linguistics",
    month = oct,
    year = "2022",
    address = "Gyeongju, Republic of Korea",
    publisher = "International Committee on Computational Linguistics",
    url = "https://aclanthology.org/2022.coling-1.606",
    pages = "6955--6965",
}

Requirements

• CUDA：11.4
• Python 3
• PyTorch 1.10.2

The code has been tested on Ubuntu 20.04.3 LTS using a single 3090(24G).

Quick Start

1. Download the pertrained "BERT-Base, Chinese" model. And then put the model file pytorch_model.bin to the folder src/bert-base-chinese

2. Run our model A^2Net.

   • python src/main.py