@inproceedings{yu-etal-2022-cocolm,
    title = "{C}o{C}o{LM}: Complex Commonsense Enhanced Language Model with Discourse Relations",
    author = "Yu, Changlong  and
      Zhang, Hongming  and
      Song, Yangqiu  and
      Ng, Wilfred",
    editor = "Muresan, Smaranda  and
      Nakov, Preslav  and
      Villavicencio, Aline",
    booktitle = "Findings of the Association for Computational Linguistics: ACL 2022",
    month = may,
    year = "2022",
    address = "Dublin, Ireland",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2022.findings-acl.93/",
    doi = "10.18653/v1/2022.findings-acl.93",
    pages = "1175--1187",
    abstract = "Large-scale pre-trained language models have demonstrated strong knowledge representation ability. However, recent studies suggest that even though these giant models contain rich simple commonsense knowledge (e.g., bird can fly and fish can swim.), they often struggle with complex commonsense knowledge that involves multiple eventualities (verb-centric phrases, e.g., identifying the relationship between ``Jim yells at Bob'' and ``Bob is upset''). To address this issue, in this paper, we propose to help pre-trained language models better incorporate complex commonsense knowledge. Unlike direct fine-tuning approaches, we do not focus on a specific task and instead propose a general language model named CoCoLM. Through the careful training over a large-scale eventuality knowledge graph ASER, we successfully teach pre-trained language models (i.e., BERT and RoBERTa) rich multi-hop commonsense knowledge among eventualities. Experiments on multiple commonsense tasks that require the correct understanding of eventualities demonstrate the effectiveness of CoCoLM."
}