Our extracted DimonGen dataset is vailable at here: dataset.
Python 3.9.12, Pytorch 1.11.0, Hugging Face Transformers 4.19.4
download the external knowledge documents from here, and put the files into data folder, and then run
python retrieval/initial_retriever.py
python retrieval/mixture_of_scorer.py
Set the default --method_name moree in main.py then run
python main.py
To run baseline methods, you need to set the method_name to different baseline methods.
@article{liu2022dimongen,
title={DimonGen: Diversified Generative Commonsense Reasoning for Explaining Concept Relationships},
author={Liu, Chenzhengyi and Huang, Jie and Zhu, Kerui and Chang, Kevin Chen-Chuan},
journal={arXiv preprint arXiv:2212.10545},
year={2022}
}