A Survey on Explainability in Machine Reading Comprehension

TL;DR

This survey reviews the development, evaluation, and challenges of explainability in Machine Reading Comprehension, highlighting key benchmarks, methodologies, and future research directions in the field.

Contribution

It provides a comprehensive overview of explainability approaches, evaluation methods, and open questions in MRC, offering a structured understanding of progress and gaps.

Findings

Evolution of representation and inference challenges

Assessment methodologies for explainable systems

Identification of open research questions

Abstract

This paper presents a systematic review of benchmarks and approaches for explainability in Machine Reading Comprehension (MRC). We present how the representation and inference challenges evolved and the steps which were taken to tackle these challenges. We also present the evaluation methodologies to assess the performance of explainable systems. In addition, we identify persisting open research questions and highlight critical directions for future work.