Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/88898
Title: A conditional variational framework for dialog generation
Authors: Shen, XY
Su, H
Li, YR 
Li, WJ 
Niu, SZ
Zhao, Y
Aizawa, A
Long, GP
Issue Date: 2017
Source: Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, July 2017, Vancouver, Canada, v. 2, p. 504-509
Abstract: Deep latent variable models have been shown to facilitate the response generation for open-domain dialog systems. However, these latent variables are highly randomized, leading to uncontrollable generated responses. In this paper, we propose a framework allowing conditional response generation based on specific attributes. These attributes can be either manually assigned or automatically detected. Moreover, the dialog states for both speakers are modeled separately in order to reflect personal features. We validate this framework on two different scenarios, where the attribute refers to genericness and sentiment states respectively. The experiment result testified the potential of our model, where meaningful responses can be generated in accordance with the specified attributes.
Publisher: Association for Computational Linguistics
DOI: 10.18653/v1/P17-2080
Rights: ©2017 Association for Computational Linguistics
Creative Commons 4.0 BY (Attribution) license (https://creativecommons.org/licenses/by/4.0/)
The following publication Shen, X. Y., Su, H., Li, Y. R., Li, W. J., Niu, S. Z., Zhao, Y., . . . Long, G. P. (2017). A conditional variational framework for dialog generation. Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, July 2017, Vancouver, Canada, v. 2, p. 504-509, 504-509 is available at https://dx.doi.org/10.18653/v1/P17-2080
Appears in Collections:Conference Paper

Files in This Item:
File Description SizeFormat 
Shen_Conditional_Variational_Framework.pdf236.8 kBAdobe PDFView/Open
Access
View full-text via PolyU eLinks SFX Query
Show full item record
PIRA download icon_1.1View/Download Full Text

WEB OF SCIENCETM
Citations

18
Citations as of Jan 21, 2021

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.