Weighted Global Normalization for Multiple Choice Reading Comprehension over Long Documents

Abstract

Motivated by recent evidence pointing out the fragility of high-performing span prediction models, we direct our attention to multiple choice reading comprehension. In particular, this work introduces a novel method for improving answer selection on long documents through weighted global normalization of predictions over portions of the documents. We show that applying our method to a span prediction model adapted for answer selection helps model performance on long summaries from NarrativeQA, a challenging reading comprehension dataset with an answer selection task, and we strongly improve on the task baseline performance by +36.2 Mean Reciprocal Rank.

References in corpus (4)

• QANet: Combining Local Convolution with Global Self-Attention for Reading Comprehension
• Adversarial Examples for Evaluating Reading Comprehension Systems
• Reading Wikipedia to Answer Open-Domain Questions
• Yuanfudao at SemEval-2018 Task 11: Three-way Attention and Relational Knowledge for Commonsense Machine Comprehension