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

Dialogue state tracking (DST) aims to convert the dialogue history intodialogue states which consist of slot-value pairs. As condensed structuralinformation memorizing all history information, the dialogue state in the lastturn is typically adopted as the input for predicting the current state by DSTmodels. However, these models tend to keep the predicted slot values unchanged,which is defined as state momentum in this paper. Specifically, the modelsstruggle to update slot values that need to be changed and correct wronglypredicted slot values in the last turn. To this end, we propose MoNET to tacklestate momentum via noise-enhanced training. First, the previous state of eachturn in the training data is noised via replacing some of its slot values.Then, the noised previous state is used as the input to learn to predict thecurrent state, improving the model's ability to update and correct slot values.Furthermore, a contrastive context matching framework is designed to narrow therepresentation distance between a state and its corresponding noised variant,which reduces the impact of noised state and makes the model better understandthe dialogue history. Experimental results on MultiWOZ datasets show that MoNEToutperforms previous DST methods. Ablations and analysis verify theeffectiveness of MoNET in alleviating state momentum and improving anti-noiseability.

MoNET: Tackle State Momentum via Noise-Enhanced Training for Dialogue State Tracking