SAE

Annotated data and code for our COLING 2022 paper: A Structure-Aware Argument Encoder for Literature Discourse Analysis.

Files and Folders

• data folder：Data (For details, see "Data.zip". Directly copy all files in the "data" folder in the .zip file)

• log folder：Log file during traning

• model folder

  • BERT_LSTM_arg.py：Model code (SAE, SE)
  • BERT_arg.py：Model code (p-SAE)

• output folder：Save the best model on the test set during training

• preds文件夹：Save predict result in zero-shot transfer learning

• pretrained_model folder：Pretrained model (Download "bert-base-uncased" (https://huggingface.co/bert-base-uncased/tree/main) and put it here)

• util folder：Utility function code

  • config.py：Configuration file to get command line parameters
  • measure.py：Define performance metrics (accuracy, precision, recall, F1) and print metrics reports

• train.py：Training code

• train_inference.py：Code for inference only, used for zero-shot transfer learning, predict results saved in the "preds" folder

• data_pre.py：Daraloader code

• command.md：Command line example file

Experiment Environment

• cuda10.1

  • python 3.8.5, cudatoolkit: 10.1, torch: 1.7.1, transformer, nltk, scikit-learn
  • pip install torch==1.7.1+cu101 torchvision==0.8.2+cu101 torchaudio==0.7.2 -f https://download.pytorch.org/whl/torch_stable.html

• cuda11.1

  • python 3.8.5, cudatoolkit: 11.1, torch: 1.8.1, transformer, nltk, scikit-learn

  • pip3 install torch==1.8.1+cu111 torchvision==0.9.1+cu111 torchaudio===0.8.1 -f https://download.pytorch.org/whl/torch_stable.html

Get Started

You can try command lines in the "command.md" file.

Annotation Tool

The annotation tool is available at https://github.com/lemuria-wchen/amcl.

How to Cite

If you extend or use this work, please cite the paper where it was introduced.

@inproceedings{li-etal-2022-structure,
    title = "A Structure-Aware Argument Encoder for Literature Discourse Analysis",
    author = "Li, Yinzi  and
      Chen, Wei  and
      Wei, Zhongyu  and
      Huang, Yujun  and
      Wang, Chujun  and
      Wang, Siyuan  and
      Zhang, Qi  and
      Huang, Xuanjing  and
      Wu, Libo",
    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-1.619",
    pages = "7093--7098",
    abstract = "Existing research for argument representation learning mainly treats tokens in the sentence equally and ignores the implied structure information of argumentative context. In this paper, we propose to separate tokens into two groups, namely framing tokens and topic ones, to capture structural information of arguments. In addition, we consider high-level structure by incorporating paragraph-level position information. A novel structure-aware argument encoder is proposed for literature discourse analysis. Experimental results on both a self-constructed corpus and a public corpus show the effectiveness of our model. Resources are available at https://github.com/lemuria-wchen/SAE.",
}