Extract, Integrate, Compete: Towards Verification Style Reading Comprehension

TL;DR

This paper introduces VGaokao, a challenging Chinese reading comprehension dataset for native speakers, and proposes an Extract-Integrate-Compete method that enhances understanding through evidence selection and pairwise comparison.

Contribution

The paper presents a new verification-style dataset for native Chinese speakers and a novel approach that improves comprehension by iterative evidence extraction and competitive learning.

Findings

Our method outperforms baselines on VGaokao.

The approach is efficient and explainable.

The dataset and code are publicly released.

Abstract

In this paper, we present a new verification style reading comprehension dataset named VGaokao from Chinese Language tests of Gaokao. Different from existing efforts, the new dataset is originally designed for native speakers' evaluation, thus requiring more advanced language understanding skills. To address the challenges in VGaokao, we propose a novel Extract-Integrate-Compete approach, which iteratively selects complementary evidence with a novel query updating mechanism and adaptively distills supportive evidence, followed by a pairwise competition to push models to learn the subtle difference among similar text pieces. Experiments show that our methods outperform various baselines on VGaokao with retrieved complementary evidence, while having the merits of efficiency and explainability. Our dataset and code are released for further research.