Incorporating Causal Analysis into Diversified and Logical Response Generation
This work addresses the challenge of improving response quality in dialogue systems for applications like chatbots, though it is incremental as it builds on existing CVAE models.
The paper tackled the problem of generating diversified but relevant and logical dialogue responses by incorporating causal analysis to mitigate confounding bias and preserve relevant information, resulting in a model that outperforms state-of-the-art methods in automatic metrics and human evaluations.
Although the Conditional Variational AutoEncoder (CVAE) model can generate more diversified responses than the traditional Seq2Seq model, the responses often have low relevance with the input words or are illogical with the question. A causal analysis is carried out to study the reasons behind, and a methodology of searching for the mediators and mitigating the confounding bias in dialogues is provided. Specifically, we propose to predict the mediators to preserve relevant information and auto-regressively incorporate the mediators into generating process. Besides, a dynamic topic graph guided conditional variational autoencoder (TGG-CVAE) model is utilized to complement the semantic space and reduce the confounding bias in responses. Extensive experiments demonstrate that the proposed model is able to generate both relevant and informative responses, and outperforms the state-of-the-art in terms of automatic metrics and human evaluations.