End-to-End Answer Chunk Extraction and Ranking for Reading Comprehension

TL;DR

This paper introduces DCR, an end-to-end neural reading comprehension model that extracts and ranks variable-length answer chunks from documents, achieving state-of-the-art results on SQuAD.

Contribution

The paper presents a novel neural RC model capable of extracting and ranking answer chunks of variable length, unlike previous models limited to single tokens or entities.

Findings

Achieves state-of-the-art exact match scores on SQuAD.

Outperforms previous models in F1 score.

Effectively predicts variable-length answer spans.

Abstract

This paper proposes dynamic chunk reader (DCR), an end-to-end neural reading comprehension (RC) model that is able to extract and rank a set of answer candidates from a given document to answer questions. DCR is able to predict answers of variable lengths, whereas previous neural RC models primarily focused on predicting single tokens or entities. DCR encodes a document and an input question with recurrent neural networks, and then applies a word-by-word attention mechanism to acquire question-aware representations for the document, followed by the generation of chunk representations and a ranking module to propose the top-ranked chunk as the answer. Experimental results show that DCR achieves state-of-the-art exact match and F1 scores on the SQuAD dataset.