Open-Ended Commonsense Reasoning with Unrestricted Answer Scope

Publication Date: 12/10/2023

Event: The 2023 Conference on Empirical Methods in Natural Language Processing (EMNLP 2023), Singapore

Reference: pp. 8035-8047, 2023

Authors: Chen Ling, Emory University; Xuchao Zhang, Microsoft; Xujiang Zhao, NEC Laboratories America, Inc.; Yifeng Wu, Emory University; Yanchi Liu, NEC Laboratories America, Inc.; Wei Cheng, NEC Laboratories America, Inc.; Takao Osaki, NEC Corporation; Katsushi Matsuda, NEC Corporation; Haifeng Chen, NEC Laboratories America, Inc.; Liang Zhao, Emory University

Abstract: Open-ended Commonsense Reasoning is defined as solving a commonsense question without providing 1) a short list of answer candidates and 2) a pre-defined answer scope. Conventional ways of formulating the commonsense question into a question-answering form or utilizing external knowledge to learn retrieval-based methods are less applicable in the open-ended setting due to an inherent challenge. Without pre-defining an answer scope or a few candidates, open-ended commonsense reasoning entails predicting answers by searching over an extremely large searching space. Moreover, most questions require implicit multi-hop reasoning, which presents even more challenges to our problem. In this work, we leverage pre-trained language models to iteratively retrieve reasoning paths on the external knowledge base, which does not require task-specific supervision. The reasoning paths can help to identify the most precise answer to the commonsense question. We conduct experiments on two commonsense benchmark datasets. Compared to other approaches, our proposed method achieves better performance both quantitatively and qualitatively.

Publication Link: