Improving Machine Reading Comprehension with Single-choice Decision and Transfer Learning
This work addresses a domain-specific problem for researchers in NLP by improving multi-choice MRC performance, though it is incremental as it builds on existing models like ALBERT.
The paper tackles the challenge of transferring knowledge from other machine reading comprehension tasks to multi-choice MRC by reconstructing it as a single-choice binary classification problem, achieving new state-of-the-art results on the RACE dataset.
Multi-choice Machine Reading Comprehension (MMRC) aims to select the correct answer from a set of options based on a given passage and question. Due to task specific of MMRC, it is non-trivial to transfer knowledge from other MRC tasks such as SQuAD, Dream. In this paper, we simply reconstruct multi-choice to single-choice by training a binary classification to distinguish whether a certain answer is correct. Then select the option with the highest confidence score. We construct our model upon ALBERT-xxlarge model and estimate it on the RACE dataset. During training, We adopt AutoML strategy to tune better parameters. Experimental results show that the single-choice is better than multi-choice. In addition, by transferring knowledge from other kinds of MRC tasks, our model achieves a new state-of-the-art results in both single and ensemble settings.