Read + Verify: Machine Reading Comprehension with Unanswerable Questions

Machine reading comprehension with unanswerable questions aims to abstain from answering when no answer can be inferred. In addition to extract answers, previous works usually predict an additional "no-answer" probability to detect unanswerable cases. However, they fail to validate the answerability of questions by verifying the legitimacy of predicted answers. To address this problem, we propose a novel read-then-verify system, which not only utilizes a neural reader to extract candidate answers and produce no-answer probabilities, but also leverages an answer verifier to validate whether the predicted answer is entailed by the input snippets. Moreover, we introduce two auxiliary losses to help the reader better handle answer extraction and no-answer detection, and investigate three different architectures for the answer verifier. Our experiments on the SQuAD 2.0 dataset show that our system can achieve a score of 74.2 F1 on the test set, outperforming all previous approaches at the time of submission (Aug. 23th, 2018).