Extract, Integrate, Compete: Towards Verification Style Reading Comprehension
This work addresses reading comprehension challenges for native speakers in educational testing, but it is incremental as it builds on existing verification-style datasets and methods.
The authors tackled the problem of verification-style reading comprehension by creating a new dataset, VGaokao, from Chinese Gaokao tests, which requires advanced language understanding for native speakers, and proposed an Extract-Integrate-Compete approach that outperforms baselines with improved efficiency and explainability.
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.