Topic-Guided Variational Autoencoders for Text Generation
This work addresses text generation for natural language processing applications, offering an incremental improvement by integrating topic modeling into variational autoencoders.
The paper tackles the problem of generating diverse and topic-coherent text by proposing a topic-guided variational autoencoder (TGVAE) that uses a Gaussian mixture model prior based on latent topics, and it shows improved performance over existing methods in unconditional and conditional text generation tasks.
We propose a topic-guided variational autoencoder (TGVAE) model for text generation. Distinct from existing variational autoencoder (VAE) based approaches, which assume a simple Gaussian prior for the latent code, our model specifies the prior as a Gaussian mixture model (GMM) parametrized by a neural topic module. Each mixture component corresponds to a latent topic, which provides guidance to generate sentences under the topic. The neural topic module and the VAE-based neural sequence module in our model are learned jointly. In particular, a sequence of invertible Householder transformations is applied to endow the approximate posterior of the latent code with high flexibility during model inference. Experimental results show that our TGVAE outperforms alternative approaches on both unconditional and conditional text generation, which can generate semantically-meaningful sentences with various topics.