MetaWSD

This is the official code for the paper Learning to Learn to Disambiguate: Meta-Learning for Few-Shot Word Sense Disambiguation, published at Findings of EMNLP.

Getting started

• Clone the repository: git clone [email protected]:Nithin-Holla/MetaWSD.git.
• Create a virtual environment.
• Install the required packages: pip install -r MetaWSD/requirements.txt.
• Create a directory for storing the data: mkdir data.
• Navigate to the data directory: cd data.
• Clone the repository containing the data: git clone [email protected]:google-research-datasets/word_sense_disambigation_corpora.git.
• Navigate back: cd ..

Preparing the data

• The first step is to generate the sense inventory: python MetaWSD/scripts/wsd_gen_sense_inventory.py.
• Next, generate episodes from the data: python MetaWSD/scripts/generate_wsd_data.py --n_support_examples N_SUPPORT_EXAMPLES --n_query_examples N_QUERY_EXAMPLES --n_train_episodes N_TRAIN_EPISODES.

Training the models

• The YAML configuration files for all the models are in config/wsd. To train a model, run python MetaWSD/train_wsd.py --config CONFIG_FILE.
• For using the non-episodic baseline, switch to the baseline branch. For all the other models, use the code in the master branch.
• Training on multiple GPUs is supported for the MAML variants only. In order to use multiple GPUs, specify the flag --multi_gpu.

Troubleshooting

If you have a RuntimeError with Proto(FO)MAML and BERT, you can install the higher library from this fork: https://github.com/Nithin-Holla/higher, which has a temporary fix for this. Also, replace diffopt.step(loss) with diffopt.step(loss, retain_graph=True) in models/seq_meta.py.

Citation

If you use this code repository, please consider citing the paper:

@inproceedings{holla-etal-2020-learning,
    title = "Learning to Learn to Disambiguate: Meta-Learning for Few-Shot Word Sense Disambiguation",
    author = "Holla, Nithin and Mishra, Pushkar and Yannakoudakis, Helen and Shutova, Ekaterina",
    booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: Findings",
    month = nov,
    year = "2020",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://www.aclweb.org/anthology/2020.findings-emnlp.405",
    pages = "4517--4533"
}