Complementary Evidence Identification in Open-Domain Question Answering
This addresses the challenge of finding diverse and sufficient evidence for complex questions in open-domain QA, but it is incremental as it builds on existing QA frameworks.
The paper tackles the problem of efficiently identifying a small set of passages that provide comprehensive evidence from multiple aspects to answer complex open-domain questions, and the proposed method significantly improves the accuracy of complementary evidence selection.
This paper proposes a new problem of complementary evidence identification for open-domain question answering (QA). The problem aims to efficiently find a small set of passages that covers full evidence from multiple aspects as to answer a complex question. To this end, we proposes a method that learns vector representations of passages and models the sufficiency and diversity within the selected set, in addition to the relevance between the question and passages. Our experiments demonstrate that our method considers the dependence within the supporting evidence and significantly improves the accuracy of complementary evidence selection in QA domain.