Towards Less Generic Responses in Neural Conversation Models: A Statistical Re-weighting Method

Yahui Liu, Wei Bi, Jun Gao, Xiaojiang Liu, Jian Yao, Shuming Shi


Abstract
Sequence-to-sequence neural generation models have achieved promising performance on short text conversation tasks. However, they tend to generate generic/dull responses, leading to unsatisfying dialogue experience. We observe that in the conversation tasks, each query could have multiple responses, which forms a 1-to-n or m-to-n relationship in the view of the total corpus. The objective function used in standard sequence-to-sequence models will be dominated by loss terms with generic patterns. Inspired by this observation, we introduce a statistical re-weighting method that assigns different weights for the multiple responses of the same query, and trains the common neural generation model with the weights. Experimental results on a large Chinese dialogue corpus show that our method improves the acceptance rate of generated responses compared with several baseline models and significantly reduces the number of generated generic responses.
Anthology ID:
D18-1297
Volume:
Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing
Month:
October-November
Year:
2018
Address:
Brussels, Belgium
Editors:
Ellen Riloff, David Chiang, Julia Hockenmaier, Jun’ichi Tsujii
Venue:
EMNLP
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
2769–2774
Language:
URL:
https://aclanthology.org/D18-1297
DOI:
10.18653/v1/D18-1297
Bibkey:
Cite (ACL):
Yahui Liu, Wei Bi, Jun Gao, Xiaojiang Liu, Jian Yao, and Shuming Shi. 2018. Towards Less Generic Responses in Neural Conversation Models: A Statistical Re-weighting Method. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, pages 2769–2774, Brussels, Belgium. Association for Computational Linguistics.
Cite (Informal):
Towards Less Generic Responses in Neural Conversation Models: A Statistical Re-weighting Method (Liu et al., EMNLP 2018)
Copy Citation:
PDF:
https://aclanthology.org/D18-1297.pdf
Video:
 https://aclanthology.org/D18-1297.mp4
Code
 yhlleo/Reweighting