Machine Comprehension Using Match-LSTM and Answer Pointer

29 Aug 2016 Shuohang Wang Jing Jiang

Machine comprehension of text is an important problem in natural language processing. A recently released dataset, the Stanford Question Answering Dataset (SQuAD), offers a large number of real questions and their answers created by humans through crowdsourcing... (read more)

PDF Abstract

Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Question Answering SQuAD1.1 Match-LSTM with Ans-Ptr (Boundary) (ensemble) EM 67.901 # 146
F1 77.022 # 151
Question Answering SQuAD1.1 Match-LSTM with Bi-Ans-Ptr (Boundary) EM 64.744 # 156
F1 73.743 # 160
Question Answering SQuAD1.1 Match-LSTM with Ans-Ptr (Boundary) EM 60.474 # 165
F1 70.695 # 168
Question Answering SQuAD1.1 Match-LSTM with Ans-Ptr (Sentence) EM 54.505 # 168
F1 67.748 # 170
Question Answering SQuAD1.1 dev Match-LSTM with Bi-Ans-Ptr (Boundary+Search+b) EM 64.1 # 37
F1 64.7 # 41

Methods used in the Paper