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/

Global: Zendy Plus annual

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

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The capability to generate responses with diversity and faithfulness usingfactual knowledge is paramount for creating a human-like, trustworthy dialoguesystem. Common strategies either adopt a two-step paradigm, which optimizesknowledge selection and response generation separately, and may overlook theinherent correlation between these two tasks, or leverage conditionalvariational method to jointly optimize knowledge selection and responsegeneration by employing an inference network. In this paper, we present anend-to-end learning framework, termed Sequential Posterior Inference (SPI),capable of selecting knowledge and generating dialogues by approximatelysampling from the posterior distribution. Unlike other methods, SPI does notrequire the inference network or assume a simple geometry of the posteriordistribution. This straightforward and intuitive inference procedure of SPIdirectly queries the response generation model, allowing for accurate knowledgeselection and generation of faithful responses. In addition to modelingcontributions, our experimental results on two common dialogue datasets (Wizardof Wikipedia and Holl-E) demonstrate that SPI outperforms previous strongbaselines according to both automatic and human evaluation metrics.

Diverse and Faithful Knowledge-Grounded Dialogue Generation via Sequential Posterior Inference