Modeling Content-Emotion Duality via Disentanglement for Empathetic Conversation
This addresses the problem of generating more empathetic responses in conversational AI, though it appears incremental as it builds on existing disentanglement methods for a specific domain.
The authors tackled empathetic response generation by modeling content-emotion duality via disentanglement, achieving state-of-the-art performance on benchmark datasets with improvements in both automatic and human metrics.
The task of empathetic response generation aims to understand what feelings a speaker expresses on his/her experiences and then reply to the speaker appropriately. To solve the task, it is essential to model the content-emotion duality of a dialogue, which is composed of the content view (i.e., what personal experiences are described) and the emotion view (i.e., the feelings of the speaker on these experiences). To this end, we design a framework to model the Content-Emotion Duality (CEDual) via disentanglement for empathetic response generation. With disentanglement, we encode the dialogue history from both the content and emotion views, and then generate the empathetic response based on the disentangled representations, thereby both the content and emotion information of the dialogue history can be embedded in the generated response. The experiments on the benchmark dataset EMPATHETICDIALOGUES show that the CEDual model achieves state-of-the-art performance on both automatic and human metrics, and it also generates more empathetic responses than previous methods.