@inproceedings{han-etal-2023-graph,
    title = "A Graph-Guided Reasoning Approach for Open-ended Commonsense Question Answering",
    author = "Han, Zhen  and
      Feng, Yue  and
      Sun, Mingming",
    editor = "Surdeanu, Mihai  and
      Riloff, Ellen  and
      Chiticariu, Laura  and
      Frietag, Dayne  and
      Hahn-Powell, Gus  and
      Morrison, Clayton T.  and
      Noriega-Atala, Enrique  and
      Sharp, Rebecca  and
      Valenzuela-Escarcega, Marco",
    booktitle = "Proceedings of the 2nd Workshop on Pattern-based Approaches to NLP in the Age of Deep Learning",
    month = dec,
    year = "2023",
    address = "Singapore",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/fix-sig-urls/2023.pandl-1.3/",
    doi = "10.18653/v1/2023.pandl-1.3",
    pages = "20--24",
    abstract = "Recently, end-to-end trained models for multiple-choice commonsense question answering (QA) have delivered promising results. However, such question-answering systems cannot be directly applied in real-world scenarios where answer candidates are not provided. Hence, a new benchmark challenge set for open-ended commonsense reasoning (OpenCSR) has been recently released, which contains natural science questions without any predefined choices. On the OpenCSR challenge set, many questions require implicit multi-hop reasoning and have a large decision space, reflecting the difficult nature of this task. Existing work on OpenCSR sorely focuses on improving the retrieval process, which extracts relevant factual sentences from a textual knowledge base, leaving the important and non-trivial reasoning task outside the scope. In this work, we extend the scope to include a reasoner that constructs a question-dependent open knowledge graph based on retrieved supporting facts and employs a sequential subgraph reasoning process to predict the answer. The subgraph can be seen as a concise and compact graphical explanation of the prediction. Experiments on two OpenCSR datasets show that the proposed model achieves great performance on benchmark OpenCSR datasets."
}