Dev set: Top1 acc: 0.9793 Top2 acc: 1.0 Top2 coverage: 0.9669 Type acc: 0.9504 Test set: Top1 acc: 0.9842 Top2 acc: 1.0 Top2 coverage: 0.9421 Type acc: 0.9368 Top1 acc: the top1 predicted mode in golden answer modes Top2 acc: one of the top2 predicted modes in golden answer modes Top2 coverage: the ratio that the top2 predicted modes cover the golden modes