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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Dialogue state tracking (DST) plays a key role in task-oriented dialoguesystems to monitor the user's goal. In general, there are two strategies totrack a dialogue state: predicting it from scratch and updating it fromprevious state. The scratch-based strategy obtains each slot value by inquiringall the dialogue history, and the previous-based strategy relies on the currentturn dialogue to update the previous dialogue state. However, it is hard forthe scratch-based strategy to correctly track short-dependency dialogue statebecause of noise; meanwhile, the previous-based strategy is not very useful forlong-dependency dialogue state tracking. Obviously, it plays different rolesfor the context information of different granularity to track different kindsof dialogue states. Thus, in this paper, we will study and discuss how thecontext information of different granularity affects dialogue state tracking.First, we explore how greatly different granularities affect dialogue statetracking. Then, we further discuss how to combine multiple granularities fordialogue state tracking. Finally, we apply the findings about contextgranularity to few-shot learning scenario. Besides, we have publicly releasedall codes.

Comprehensive Study: How the Context Information of Different Granularity Affects Dialogue State Tracking?