Mem2Seq: Effectively Incorporating Knowledge Bases into End-to-End Task-Oriented Dialog Systems
This addresses the problem of knowledge integration in dialog systems for applications like virtual assistants, though it appears incremental as it builds on existing neural generative models.
The paper tackles the challenge of incorporating knowledge bases into end-to-end task-oriented dialog systems by proposing Mem2Seq, a model that combines multi-hop attention over memories with pointer networks, achieving state-of-the-art performance on three datasets with faster training.
End-to-end task-oriented dialog systems usually suffer from the challenge of incorporating knowledge bases. In this paper, we propose a novel yet simple end-to-end differentiable model called memory-to-sequence (Mem2Seq) to address this issue. Mem2Seq is the first neural generative model that combines the multi-hop attention over memories with the idea of pointer network. We empirically show how Mem2Seq controls each generation step, and how its multi-hop attention mechanism helps in learning correlations between memories. In addition, our model is quite general without complicated task-specific designs. As a result, we show that Mem2Seq can be trained faster and attain the state-of-the-art performance on three different task-oriented dialog datasets.