RACE: Large-scale ReAding Comprehension Dataset From Examinations

TL;DR

RACE is a large-scale dataset derived from Chinese school exams designed to evaluate reading comprehension models, emphasizing reasoning skills and highlighting the performance gap between current models and humans.

Contribution

The paper introduces RACE, a comprehensive and challenging dataset for machine reading comprehension, focusing on reasoning and real exam questions.

Findings

State-of-the-art models achieve 43% accuracy on RACE.

Human performance on RACE is approximately 95%.

RACE covers diverse topics and reasoning types.

Abstract

We present RACE, a new dataset for benchmark evaluation of methods in the reading comprehension task. Collected from the English exams for middle and high school Chinese students in the age range between 12 to 18, RACE consists of near 28,000 passages and near 100,000 questions generated by human experts (English instructors), and covers a variety of topics which are carefully designed for evaluating the students' ability in understanding and reasoning. In particular, the proportion of questions that requires reasoning is much larger in RACE than that in other benchmark datasets for reading comprehension, and there is a significant gap between the performance of the state-of-the-art models (43%) and the ceiling human performance (95%). We hope this new dataset can serve as a valuable resource for research and evaluation in machine comprehension. The dataset is freely available at http://www.cs.cmu.edu/~glai1/data/race/ and the code is available at https://github.com/qizhex/RACE_AR_baselines.