Publication Type

Conference Proceeding Article

Version

publishedVersion

Publication Date

4-2020

Abstract

In open-domain dialogue systems, dialogue cues such as emotion, persona, and emoji can be incorporated into conversation models for strengthening the semantic relevance of generated responses. Existing neural response generation models either incorporate dialogue cue into decoder’s initial state or embed the cue indiscriminately into the state of every generated word, which may cause the gradients of the embedded cue to vanish or disturb the semantic relevance of generated words during back propagation. In this paper, we propose a Cue Adaptive Decoder (CueAD) that aims to dynamically determine the involvement of a cue at each generation step in the decoding. For this purpose, we extend the Gated Recurrent Unit (GRU) network with an adaptive cue representation for facilitating cue incorporation, in which an adaptive gating unit is utilized to decide when to incorporate cue information so that the cue can provide useful clues for enhancing the semantic relevance of the generated words. Experimental results show that Cu

Keywords

dialogue generation, vanishing gradient problem, disturbing gradient problem, cue adaptive decoder

Discipline

Artificial Intelligence and Robotics | Databases and Information Systems

Research Areas

Data Science and Engineering

Publication

WWW '20: Proceedings of The Web Conference 2020, Taipei, Taiwan, April 20-24

First Page

2570

Last Page

2576

Identifier

10.1145/3366423.3380008

City or Country

Taipei, Taiwan

Additional URL

https://doi.org/10.1145/3366423.3380008

Download

Share

COinS