@inproceedings{liang-surdeanu-2020-transformers,
    title = "Do Transformers Dream of Inference, or Can Pretrained Generative Models Learn Implicit Inferential Rules?",
    author = "Liang, Zhengzhong  and
      Surdeanu, Mihai",
    editor = "Rogers, Anna  and
      Sedoc, Jo{\~a}o  and
      Rumshisky, Anna",
    booktitle = "Proceedings of the First Workshop on Insights from Negative Results in NLP",
    month = nov,
    year = "2020",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/jlcl-multiple-ingestion/2020.insights-1.12/",
    doi = "10.18653/v1/2020.insights-1.12",
    pages = "76--81",
    abstract = "Large pretrained language models (LM) have been used successfully for multi-hop question answering. However, most of these directions are not interpretable, as they do not make the inference hops necessary to explain a candidate answer explicitly. In this work, we investigate the capability of a state-of-the-art transformer LM to generate explicit inference hops, i.e., to infer a new statement necessary to answer a question given some premise input statements. Our analysis shows that such LMs can generate new statements for some simple inference types, but performance remains poor for complex, real-world inference types such as those that require monotonicity, composition, and commonsense knowledge."
}