Reference Knowledgeable Network for Machine Reading Comprehension
Yilin Zhao, Zhuosheng Zhang, Hai Zhao · IEEE/ACM Transactions on Audio Speech and Language Processing · 2022
Multi-choice Machine Reading Comprehension (MRC) as a challenge requires models to select the most appropriate answer from a set of candidates with a given passage and question. Most of the existing researches focus on the modeling of specific tasks or complex networks, without explicitly referring to relevant and credible external knowledge sources, which are supposed to greatly make up for the deficiency of the given passage. Thus we propose a novel reference-based knowledge enhancement model calledReferenceKnowledgeableNetwork (RekNet), which simulates human reading strategies to refine critical information from the passage and quote explicit knowledge in necessity. In detail,RekNetrefines fine-grained critical information and defines it asReference Span, then quotes explicit knowledge quadruples by the co-occurrence information ofReference Spanand candidates. The proposedRekNetis evaluated on three multi-choice MRC benchmarks: RACE, DREAM and Cosmos QA, obtaining consistent and remarkable performance improvement with observable statistical significance level over strong baselines. Our code is available athttps://github.com/Yilin1111/RekNet.