Context-Aware Sequence-to-Sequence Models for Conversational Systems
This addresses the need for more context-aware conversational AI systems, but it is incremental as it builds on existing seq2seq models.
The paper tackled the problem of seq2seq models lacking context from previous conversation turns in conversational systems, and proposed an RNN-based method to integrate this context, with experimental results showing feasibility and effectiveness based on human judgment.
This work proposes a novel approach based on sequence-to-sequence (seq2seq) models for context-aware conversational systems. Exist- ing seq2seq models have been shown to be good for generating natural responses in a data-driven conversational system. However, they still lack mechanisms to incorporate previous conversation turns. We investigate RNN-based methods that efficiently integrate previous turns as a context for generating responses. Overall, our experimental results based on human judgment demonstrate the feasibility and effectiveness of the proposed approach.