English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

The natural language generation domain has witnessed great success thanks toTransformer models. Although they have achieved state-of-the-art generativequality, they often neglect generative diversity. Prior attempts to tackle thisissue suffer from either low model capacity or over-complicated architectures.Some recent methods employ the VAE framework to enhance diversity, but theirlatent variables fully depend on the input context, restricting exploration ofthe latent space. In this paper, we introduce VOLTA, a framework that elevatesgenerative diversity by bridging Transformer with VAE via a more effectivecross-attention-based connection, departing from conventional embeddingconcatenation or summation. Additionally, we propose integrating InfoGAN-stylelatent codes to enable input-independent variability, further diversifying thegeneration. Moreover, our framework accommodates discrete inputs alongside itsexisting support for continuous inputs. We perform comprehensive experimentswith two types of Transformers on six datasets from three different NLG tasksto show that our approach can significantly improve generative diversity whilemaintaining generative quality.

VOLTA: Improving Generative Diversity by Variational Mutual Information Maximizing Autoencoder