Improving Semantic Control in Discrete Latent Spaces with Transformer Quantized Variational Autoencoders
This addresses the challenge of semantic control in VAEs for NLP researchers and practitioners, offering improved performance in downstream tasks like natural language and symbolic reasoning, though it appears incremental as it builds on existing VQVAE and Transformer methods.
The paper tackles the problem of achieving precise semantic control in Variational AutoEncoders (VAEs) for NLP tasks by proposing T5VQVAE, a model that combines discrete latent spaces from VQVAEs with Transformer-based VAEs to guide self-attention at the token-level, resulting in outperforming state-of-the-art models like Optimus in controllability and semantic preservation across tasks such as auto-encoding, text transfer, and inference.
Achieving precise semantic control over the latent spaces of Variational AutoEncoders (VAEs) holds significant value for downstream tasks in NLP as the underlying generative mechanisms could be better localised, explained and improved upon. Recent research, however, has struggled to achieve consistent results, primarily due to the inevitable loss of semantic information in the variational bottleneck and limited control over the decoding mechanism. To overcome these challenges, we investigate discrete latent spaces in Vector Quantized Variational AutoEncoders (VQVAEs) to improve semantic control and generation in Transformer-based VAEs. In particular, We propose T5VQVAE, a novel model that leverages the controllability of VQVAEs to guide the self-attention mechanism in T5 at the token-level, exploiting its full generalization capabilities. Experimental results indicate that T5VQVAE outperforms existing state-of-the-art VAE models, including Optimus, in terms of controllability and preservation of semantic information across different tasks such as auto-encoding of sentences and mathematical expressions, text transfer, and inference. Moreover, T5VQVAE exhibits improved inference capabilities, suggesting potential applications for downstream natural language and symbolic reasoning tasks.