Question Answering with Long Multiple-Span Answers

• DOI: 10.18653/v1/2020.findings-emnlp.342
• Publication Date: 2020-11-29T14:58:51Z
• Authors: Ming Zhu; Aman Ahuja; Da-Cheng Juan; Wei Wei; Chandan K. Reddy
• Keywords: Relevance

Answering questions in many real-world applications often requires complex and precise information excerpted from texts spanned across a long document. However, currently no such annotated dataset is publicly available, which hinders the development of neural question-answering (QA) systems. To this end, we present MASH-QA, Multiple Answer Spans Healthcare Question consumer health domain, where answers may need to be multiple, non-consecutive parts text We also propose MultiCo, architecture that able capture relevance among multiple answer spans, by using query-based contextualized sentence selection approach, for forming given question. demonstrate conventional QA models are not suitable type task perform poorly setting. Extensive experiments conducted, experimental results confirm proposed model significantly outperforms state-of-the-art multi-span