Work

Learning Discourse-level Diversity for Neural Dialog Models using Conditional Variational Autoencoders

Year: 2017

Type: conference-paper

Abstract: While recent neural encoder-decoder models have shown great promise in modeling open-domain conversations, they often generate dull and generic responses.Unlike past work that has focused on diversifying the output of the decoder at word-level to alleviate this problem, we present a novel framework based on conditional variational autoencoders that captures the discourse-level diversity in the encoder.Our model uses latent variables to learn a distribution over potential conversational intents and generates diverse responses using only greedy decoders.We have further developed a novel variant that is integrated with linguistic prior knowledge for better performance.Finally, the training procedure is improved by introducing a bag-of-word loss.Our proposed models have been validated to generate significantly more diverse responses than baseline approaches and exhibit competence in discourse-level decision-making.

Source: Annual Meeting of the Association for Computational Linguistics (ACL)

DOI: 10.18653/v1/p17-1061

Authors: Tiancheng Zhao, Ran Zhao, Maxine Eskenazi

Institution: Carnegie Mellon University

Language: English

Topic: Speech and dialogue systems

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Funder: National Science Foundation

Award: CI-NEW: Collaborative Research: DialPort: Enabling Spoken Dialog Research with Real Data