Relevant CommonSense Subgraphs for “What if...” Procedural Reasoning

Chen Zheng; Parisa Kordjamshidi

ACL 2022findings9 citations

Relevant CommonSense Subgraphs for “What if...” Procedural Reasoning

Abstract

We study the challenge of learning causal reasoning over procedural text to answer “What if...” questions when external commonsense knowledge is required. We propose a novel multi-hop graph reasoning model to 1) efficiently extract a commonsense subgraph with the most relevant information from a large knowledge graph; 2) predict the causal answer by reasoning over the representations obtained from the commonsense subgraph and the contextual interactions between the questions and context. We evaluate our model on WIQA benchmark and achieve state-of-the-art performance compared to the recent models.

BibTeX

@inproceedings{zheng-kordjamshidi-2022-relevant,
    title = "Relevant {C}ommon{S}ense Subgraphs for {\textquotedblleft}What if...{\textquotedblright} Procedural Reasoning",
    author = "Zheng, Chen  and
      Kordjamshidi, Parisa",
    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://aclanthology.org/2022.findings-acl.152/",
    doi = "10.18653/v1/2022.findings-acl.152",
    pages = "1927--1933"
}