BiLSTM - Label Attention Network (BiLSTM-LAN)

Hierarchically-Refined Label Attention Network for Sequence Labeling (EMNLP 2019)

Model Structure

The model consists of two BiLSTM-LAN layers. Each BiLSTM-LAN layer is composed of a BiLSTM encoding sublayer and a label-attention inference sublayer. In paticular, the former is the same as the BiLSTM layer in the baseline model, while the latter uses multihead attention to jointly encode information from the word representation subspace and the label representation subspace.

Requirement

• Python3
• PyTorch: 0.3

Train models

• Download data and word embedding
• Run the script:

  python main.py --learning_rate 0.01 --lr_decay 0.035 --dropout 0.5 --hidden_dim 400 --lstm_layer 3 --momentum 0.9 --whether_clip_grad True --clip_grad 5.0 \
  --train_dir 'wsj_pos/train.pos' --dev_dir 'wsj_pos/dev.pos' --test_dir 'wsj_pos/test.pos' --model_dir 'wsj_pos/' --word_emb_dir 'glove.6B.100d.txt'

Performance

Cite

Leyang Cui and Yue Zhang. 2019. Hierarchically-refined label attention network for sequence labeling.InProceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 4106–4119, Hong Kong, China. Association for Computational Linguistics.

@inproceedings{cui-zhang-2019-hierarchically,
    title = "Hierarchically-Refined Label Attention Network for Sequence Labeling",
    author = "Cui, Leyang  and
      Zhang, Yue",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)",
    month = nov,
    year = "2019",
    address = "Hong Kong, China",
    publisher = "Association for Computational Linguistics",
    url = "https://www.aclweb.org/anthology/D19-1422",
    pages = "4106--4119",
}

Acknowledgments

NCRF++