Improving Machine Reading Comprehension with Single-choice Decision and Transfer Learning
arXiv:2011.03292
Abstract
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.
References in corpus (9)
- AutoML: A Survey of the State-of-the-Art
- Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism
- SQuAD: 100,000+ Questions for Machine Comprehension of Text
- Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
- Automated Machine Learning: State-of-The-Art and Open Challenges
- RACE: Large-scale ReAding Comprehension Dataset From Examinations
- NewsQA: A Machine Comprehension Dataset
- Option Comparison Network for Multiple-choice Reading Comprehension
- DUMA: Reading Comprehension with Transposition Thinking