We observe that end-to-end memory networks (MN) trained for task-oriented dialogue, such as for recommending restaurants to a user, suffer from an out-of-vocabulary (OOV) problem -- the entities returned by the Knowledge Base (KB) may not be seen by the network at training time, making it impossible for it to use them in dialogue. We propose a Hierarchical Pointer Memory Network (HyP-MN), in which the next word may be generated from the decode vocabulary or copied from a hierarchical memory maintaining KB results and previous utterances. Evaluating over the dialog bAbI tasks, we find that HyP-MN drastically outperforms MN obtaining 12% overall accuracy gains. Further analysis reveals that MN fails completely in recommending any relevant restaurant, whereas HyP-MN recommends the best next restaurant 80% of the time.

翻译：我们注意到,受过任务导向对话培训的端到端记忆网(MN),如向用户推荐餐厅,存在一个校外问题 -- -- 知识库(KB)返回的实体在培训时可能看不到,因此无法在对话中使用这些实体。我们提议建立一个等级式指针记忆网(HyP-MN),在该网络中,下一个词可以从解码词汇中生成,或从维持KB结果和以前言论的等级式记忆中复制出来。在评估BABI对话任务时,我们发现HYP-MN大大优于MN取得12%的总体准确性收益。进一步的分析显示,MND完全没有推荐任何相关的餐厅,而HyP-MN则建议80%的时间里最好的下一家餐厅。