Advanced Conditional Variational Autoencoders (A-CVAE): Towards interpreting open-domain conversation generation via disentangling latent feature representation

TL;DR

This paper introduces a novel approach to open-domain dialogue generation using a cognitively inspired disentanglement of latent features in VAEs, improving interpretability and dialogue quality.

Contribution

It proposes a mesoscopic scale feature disentanglement method guided by prior knowledge, enhancing interpretability and quality of dialogue generation.

Findings

Higher quality dialogue generation demonstrated

Improved interpretability of latent space

New metric for evaluating dialogue interpretability

Abstract

Currently end-to-end deep learning based open-domain dialogue systems remain black box models, making it easy to generate irrelevant contents with data-driven models. Specifically, latent variables are highly entangled with different semantics in the latent space due to the lack of priori knowledge to guide the training. To address this problem, this paper proposes to harness the generative model with a priori knowledge through a cognitive approach involving mesoscopic scale feature disentanglement. Particularly, the model integrates the macro-level guided-category knowledge and micro-level open-domain dialogue data for the training, leveraging the priori knowledge into the latent space, which enables the model to disentangle the latent variables within the mesoscopic scale. Besides, we propose a new metric for open-domain dialogues, which can objectively evaluate the interpretability of the latent space distribution. Finally, we validate our model on different datasets and experimentally demonstrate that our model is able to generate higher quality and more interpretable dialogues than other models.