Neural Response Ranking for Social Conversation: A Data-Efficient Approach

11/02/2018
by   Igor Shalyminov, et al.
0

The overall objective of 'social' dialogue systems is to support engaging, entertaining, and lengthy conversations on a wide variety of topics, including social chit-chat. Apart from raw dialogue data, user-provided ratings are the most common signal used to train such systems to produce engaging responses. In this paper we show that social dialogue systems can be trained effectively from raw unannotated data. Using a dataset of real conversations collected in the 2017 Alexa Prize challenge, we developed a neural ranker for selecting 'good' system responses to user utterances, i.e. responses which are likely to lead to long and engaging conversations. We show that (1) our neural ranker consistently outperforms several strong baselines when trained to optimise for user ratings; (2) when trained on larger amounts of data and only using conversation length as the objective, the ranker performs better than the one trained using ratings -- ultimately reaching a Precision@1 of 0.87. This advance will make data collection for social conversational agents simpler and less expensive in the future.

READ FULL TEXT

page 1

page 2

page 3

page 4

10/21/2021

Modeling Performance in Open-Domain Dialogue with PARADISE

There has recently been an explosion of work on spoken dialogue systems,...
05/06/2020

Weakly-Supervised Neural Response Selection from an Ensemble of Task-Specialised Dialogue Agents

Dialogue engines that incorporate different types of agents to converse ...
10/04/2019

Classification As Decoder: Trading Flexibility For Control In Neural Dialogue

Generative seq2seq dialogue systems are trained to predict the next word...
06/01/2022

Understanding How People Rate Their Conversations

User ratings play a significant role in spoken dialogue systems. Typical...
01/16/2019

Learning from Dialogue after Deployment: Feed Yourself, Chatbot!

The majority of conversations a dialogue agent sees over its lifetime oc...
11/16/2019

Classification as Decoder: Trading Flexibility for Control in Medical Dialogue

Generative seq2seq dialogue systems are trained to predict the next word...
12/20/2017

An Ensemble Model with Ranking for Social Dialogue

Open-domain social dialogue is one of the long-standing goals of Artific...