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

This paper presents a novel open-domain dialogue generation model emphasizingthe differentiation of speakers in multi-turn conversations. Differing fromprior work that solely relies on the content of conversation history togenerate a response, we argue that capturing relative social relations amongutterances (i.e., generated by either the same speaker or different persons)benefits the machine capturing fine-grained context information from aconversation history to improve context coherence in the generated response.Given that, we propose a speaker-aware Parallel Hierarchical AttentiveEncoder-Decoder (PHAED) model that aims to model each utterance with theawareness of its speaker and contextual associations with the same speaker'sprevious messages. Specifically, in a conversation involving two speakers, weregard the utterances from one speaker as responses and those from the other asqueries. After understanding queries via our encoder with inner-query andinter-query encodings, our decoder reuses the hidden states of previouslygenerated responses, instead of reconstructing these by the encoder, togenerate a new response. Our empirical results show that PHAED outperforms thestate-of-the-art in both automatic and human evaluations. Furthermore, ourablation study shows that dialogue models with speaker tokens can generallydecrease the possibility of generating non-coherent responses regarding theconversation context.

A Speaker-aware Parallel Hierarchical Attentive Encoder-Decoder Model for Multi-turn Dialogue Generation