Comparative Analysis of Neural QA models on SQuAD

TL;DR

This paper compares various neural question answering models on the SQuAD dataset, analyzing their strengths, biases, and error patterns to understand their capabilities and limitations.

Contribution

It provides a comprehensive analysis of existing neural QA models on SQuAD, highlighting model-specific biases and error characteristics for better understanding.

Findings

Models exhibit distinct biases affecting their predictions.

Error analysis reveals common challenges across models.

Quantitative and qualitative insights into model performance.

Abstract

The task of Question Answering has gained prominence in the past few decades for testing the ability of machines to understand natural language. Large datasets for Machine Reading have led to the development of neural models that cater to deeper language understanding compared to information retrieval tasks. Different components in these neural architectures are intended to tackle different challenges. As a first step towards achieving generalization across multiple domains, we attempt to understand and compare the peculiarities of existing end-to-end neural models on the Stanford Question Answering Dataset (SQuAD) by performing quantitative as well as qualitative analysis of the results attained by each of them. We observed that prediction errors reflect certain model-specific biases, which we further discuss in this paper.