Efficient Natural Language Response Suggestion for Smart Reply
This work addresses the need for efficient response suggestion in real-world applications like e-mail, though it is incremental as it builds on existing neural network techniques.
The paper tackles the problem of computationally efficient natural language response suggestion by proposing a feed-forward neural network method with n-gram embeddings, which achieves the same quality as a sequence-to-sequence approach at a significantly reduced computational cost and latency in a large-scale commercial e-mail application.
This paper presents a computationally efficient machine-learned method for natural language response suggestion. Feed-forward neural networks using n-gram embedding features encode messages into vectors which are optimized to give message-response pairs a high dot-product value. An optimized search finds response suggestions. The method is evaluated in a large-scale commercial e-mail application, Inbox by Gmail. Compared to a sequence-to-sequence approach, the new system achieves the same quality at a small fraction of the computational requirements and latency.