Search

Machine comprehension of texts often requires external common knowledge and coreference resolution in the passage. However, most current machine reading comprehension models only incorporate external common knowledge. We propose CoSp model, which incorporates both common knowledge and specific entity linking knowledge for machine reading comprehension. It employs an attention mechanism to adaptively select relevant commonsense and lexical common knowledge from knowledge bases, then it leverages the relational-GCN for reasoning on the entity graph, which is constructed by the entity coreference and co-occurrence for each passage. Hence we obtain knowledge-aware and coreference-aware contextual word representation for answer prediction. Experimental results indicate that CoSp model offers significant and consistent improvements over BERT, outperforming competitive knowledge-aware models on ReCoRD and SQuAD1.1 benchmarks.

Machine reading comprehension has been an elusive challenge in the study of machine intelligence in recent years. The cloze-style reading comprehension is a query containing a placeholder that should be replaced with an appropriate word based on reading and understanding of a related document. This paper presents a novel hybrid encoding-based model for the cloze-style Chinese machine reading comprehension task, which resembles the human brain reading and reasoning process. The model is evaluated with the Children’s Fairy Tale corpus used in The First Evaluation Workshop on Chinese Machine Reading Comprehension (CMRC2017). With an accuracy of 78.7% on the test set, this model substantially outperforms the comparison models published in CMRC2017. Also, we conduct extensive experiments on People's Daily Online, People's Daily, and Reference News datasets, and the experimental results dramatically improve a stable baseline for the Chinese machine reading comprehension task.