Machine Comprehension Using Match-LSTM and Answer Pointer

Benchmark Model Rank Results
question-answering-on-squad11Match-LSTM with Ans-Ptr (Boundary) (ensemble)#38EM: 67.901F1: 77.022
question-answering-on-squad11Match-LSTM with Bi-Ans-Ptr (Boundary)#41EM: 64.744F1: 73.743
question-answering-on-squad11Match-LSTM with Ans-Ptr (Boundary)#44EM: 60.474F1: 70.695
question-answering-on-squad11Match-LSTM with Ans-Ptr (Sentence)#46EM: 54.505F1: 67.748
question-answering-on-squad11-devMatch-LSTM with Bi-Ans-Ptr (Boundary+Search+b)#39EM: 64.1F1: 64.7