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

Recently, various neural models for multi-party conversation (MPC) haveachieved impressive improvements on a variety of tasks such as addresseerecognition, speaker identification and response prediction. However, theseexisting methods on MPC usually represent interlocutors and utterancesindividually and ignore the inherent complicated structure in MPC which mayprovide crucial interlocutor and utterance semantics and would enhance theconversation understanding process. To this end, we present MPC-BERT, apre-trained model for MPC understanding that considers learning who says whatto whom in a unified model with several elaborated self-supervised tasks.Particularly, these tasks can be generally categorized into (1) interlocutorstructure modeling including reply-to utterance recognition, identical speakersearching and pointer consistency distinction, and (2) utterance semanticsmodeling including masked shared utterance restoration and shared nodedetection. We evaluate MPC-BERT on three downstream tasks including addresseerecognition, speaker identification and response selection. Experimentalresults show that MPC-BERT outperforms previous methods by large margins andachieves new state-of-the-art performance on all three downstream tasks at twobenchmarks.

MPC-BERT: A Pre-Trained Language Model for Multi-Party Conversation Understanding