Contextual and Semantic Fusion Network for Multiple-Choice Reading Comprehension

Abstract

Multiple-choice reading comprehension (MCRC) aims to build an intelligent system that automatically selects an answer from a candidate set when given a passage and a question. Existing MCRC systems rarely consider incorporating external knowledge such as explicit semantic information. In this work, we propose a Contextual and Semantic Fusion Network (CSFN) which effectively integrates contextual and semantic representation. CSFN introduces explicit structured semantics from pre-trained semantic role labeling. Specially, we regard explicit semantic representation as an important feature to fuse with contextual representation, which enriches the representation of sentences. By combining with the transfer learning strategy, the CSFN model has better generalization over limited datasets. To evaluate the ability of our model, we conduct experiments on three MCRC benchmark datasets: RACE, DREAM, and MCTest. Experimental results demonstrate the effectiveness of our proposed model.

SSIN: Sentence Semantic Interaction Network for Multi-Choice Reading Comprehension

“…as auxiliary knowledge in addition to information from the text. Present-day scholars are also investigating the introduction of external knowledge into machine reading comprehension; for example, Jiang et al [27] in 2020 propose the use of external knowledge in the form of triads and corpora, Yang et al [28] in 2020 propose a model for answering questions by searching for external knowledge, Duan et al [29] in 2021 propose a model that can combine external knowledge and contextual fusion network, and Van et al [30] proposed a model to analyze questions using external knowledge in 2020, all of these models using external knowledge have achieved good performance, which also shows the direction for our future research. In our subsequent work, we will focus on how to effectively incorporate external knowledge into machine reading comprehension and, in response to the importance of sentence semantic vectors for reading comprehension, we will continue to investigate how to convert word vectors of articles into sentence vectors.…”