Relation/Entity-Centric Reading Comprehension

TL;DR

This paper explores entity and relation-focused reading comprehension to improve machine understanding of human language, specifically targeting question answering tasks that measure comprehension by analyzing semantic structures.

Contribution

It introduces a focus on entities and relations for reading comprehension, emphasizing their importance in understanding natural language semantics.

Findings

Enhanced understanding of entity-relation structures improves question answering accuracy.

Proposed methods better capture semantic information for comprehension tasks.

Results demonstrate improved performance on benchmark datasets.

Abstract

Constructing a machine that understands human language is one of the most elusive and long-standing challenges in artificial intelligence. This thesis addresses this challenge through studies of reading comprehension with a focus on understanding entities and their relationships. More specifically, we focus on question answering tasks designed to measure reading comprehension. We focus on entities and relations because they are typically used to represent the semantics of natural language.