Learning to Paraphrase for Question Answering
This addresses the problem of linguistic variability in question answering for users of QA systems, but it is incremental as it builds on existing paraphrase and QA methods.
The paper tackles the sensitivity of question answering systems to varied natural language expressions by learning felicitous paraphrases to capture this knowledge, resulting in consistent performance improvements across three datasets with competitive results using simple QA models.
Question answering (QA) systems are sensitive to the many different ways natural language expresses the same information need. In this paper we turn to paraphrases as a means of capturing this knowledge and present a general framework which learns felicitous paraphrases for various QA tasks. Our method is trained end-to-end using question-answer pairs as a supervision signal. A question and its paraphrases serve as input to a neural scoring model which assigns higher weights to linguistic expressions most likely to yield correct answers. We evaluate our approach on QA over Freebase and answer sentence selection. Experimental results on three datasets show that our framework consistently improves performance, achieving competitive results despite the use of simple QA models.