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

Knowledge-based dialogue systems with internet retrieval have recentlyattracted considerable attention from researchers. The dialogue systemsovercome a major limitation of traditional knowledge dialogue systems, wherethe timeliness of knowledge cannot be assured, hence providing greaterpractical application value. Knowledge-based dialogue systems with internetretrieval can be typically segmented into three tasks: Retrieval Decision,Query Generation, and Response Generation. However, many of studies assumedthat all conversations require external knowledge to continue, neglecting thecritical step of determining when retrieval is necessary. This assumption oftenleads to an over-dependence on external knowledge, even when it may not berequired. Our work addresses this oversight by employing a single unified modelfacilitated by prompt and multi-task learning approaches. This model not onlydecides whether retrieval is necessary but also generates retrieval queries andresponses. By integrating these functions, our system leverages the fullpotential of pre-trained models and reduces the complexity and costs associatedwith deploying multiple models. We conducted extensive experiments toinvestigate the mutual enhancement among the three tasks in our system. What ismore, the experiment results on the Wizint and Dusinc datasets not onlydemonstrate that our unified model surpasses the baseline performance forindividual tasks, but also reveal that it achieves comparable results whencontrasted with SOTA systems that deploy separate, specialized models for eachtask.

UniRQR: A Unified Model for Retrieval Decision, Query, and Response Generation in Internet-Based Knowledge Dialogue Systems