SQuAD: 100,000+ Questions for Machine Comprehension of Text

TL;DR

The paper introduces SQuAD, a large-scale reading comprehension dataset with over 100,000 questions from Wikipedia, enabling the development and evaluation of machine reading models.

Contribution

It provides a new, extensive dataset for machine comprehension, along with analysis of question types and baseline model performance.

Findings

Baseline model achieves 51.0% F1 score.

Human performance is 86.8%, highlighting the dataset's challenge.

Analysis reveals diverse reasoning types needed for answering questions.

Abstract

We present the Stanford Question Answering Dataset (SQuAD), a new reading comprehension dataset consisting of 100,000+ questions posed by crowdworkers on a set of Wikipedia articles, where the answer to each question is a segment of text from the corresponding reading passage. We analyze the dataset to understand the types of reasoning required to answer the questions, leaning heavily on dependency and constituency trees. We build a strong logistic regression model, which achieves an F1 score of 51.0%, a significant improvement over a simple baseline (20%). However, human performance (86.8%) is much higher, indicating that the dataset presents a good challenge problem for future research. The dataset is freely available at https://stanford-qa.com