Synonym Knowledge Enhanced Reader for Chinese Idiom Reading Comprehension

TL;DR

This paper introduces a novel Chinese idiom reading comprehension model that leverages synonym knowledge and graph neural networks to better understand idioms' meanings, achieving state-of-the-art results.

Contribution

It proposes a synonym knowledge enhanced reader that encodes idiom relationships via a graph attention network, addressing semantic-literal meaning gaps in Chinese idiom comprehension.

Findings

Achieves state-of-the-art performance on ChID dataset.

Shows that synonym relationships improve idiom understanding.

Validates the effectiveness of graph-based encoding for idiom semantics.

Abstract

Machine reading comprehension (MRC) is the task that asks a machine to answer questions based on a given context. For Chinese MRC, due to the non-literal and non-compositional semantic characteristics, Chinese idioms pose unique challenges for machines to understand. Previous studies tend to treat idioms separately without fully exploiting the relationship among them. In this paper, we first define the concept of literal meaning coverage to measure the consistency between semantics and literal meanings for Chinese idioms. With the definition, we prove that the literal meanings of many idioms are far from their semantics, and we also verify that the synonymic relationship can mitigate this inconsistency, which would be beneficial for idiom comprehension. Furthermore, to fully utilize the synonymic relationship, we propose the synonym knowledge enhanced reader. Specifically, for each idiom, we first construct a synonym graph according to the annotations from a high-quality synonym dictionary or the cosine similarity between the pre-trained idiom embeddings and then incorporate the graph attention network and gate mechanism to encode the graph. Experimental results on ChID, a large-scale Chinese idiom reading comprehension dataset, show that our model achieves state-of-the-art performance.