HL-EncDec: A Hybrid-Level Encoder-Decoder for Neural Response Generation

Sixing Wu, Dawei Zhang, Ying Li, Xing Xie, Zhonghai Wu

Abstract
Recent years have witnessed a surge of interest on response generation for neural conversation systems. Most existing models are implemented by following the Encoder-Decoder framework and operate sentences of conversations at word-level. The word-level model is suffering from the Unknown Words Issue and the Preference Issue, which seriously impact the quality of generated responses, for example, generated responses may become irrelevant or too general (i.e. safe responses). To address these issues, this paper proposes a hybrid-level Encoder-Decoder model (HL-EncDec), which not only utilizes the word-level features but also character-level features. We conduct several experiments to evaluate HL-EncDec on a Chinese corpus, experimental results show our model significantly outperforms other non-word-level models in automatic metrics and human annotations and is able to generate more informative responses. We also conduct experiments with a small-scale English dataset to show the generalization ability.
Anthology ID:
C18-1072
Volume:
Proceedings of the 27th International Conference on Computational Linguistics
Month:
August
Year:
2018
Address:
Santa Fe, New Mexico, USA
Venue:
COLING
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
845–856
Language:
URL:
https://aclanthology.org/C18-1072
DOI:
Bibkey:
Cite (ACL):
Sixing Wu, Dawei Zhang, Ying Li, Xing Xie, and Zhonghai Wu. 2018. HL-EncDec: A Hybrid-Level Encoder-Decoder for Neural Response Generation. In Proceedings of the 27th International Conference on Computational Linguistics, pages 845–856, Santa Fe, New Mexico, USA. Association for Computational Linguistics.
Cite (Informal):
HL-EncDec: A Hybrid-Level Encoder-Decoder for Neural Response Generation (Wu et al., COLING 2018)
Copy Citation:
PDF:
https://preview.aclanthology.org/paclic-22-ingestion/C18-1072.pdf
PDF Search