Learning to Ask Conversational Questions by Optimizing Levenshtein Distance

Zhongkun Liu, Pengjie Ren, Zhumin Chen, Zhaochun Ren, Maarten de Rijke, Ming Zhou


Abstract
Conversational Question Simplification (CQS) aims to simplify self-contained questions into conversational ones by incorporating some conversational characteristics, e.g., anaphora and ellipsis. Existing maximum likelihood estimation based methods often get trapped in easily learned tokens as all tokens are treated equally during training. In this work, we introduce a Reinforcement Iterative Sequence Editing (RISE) framework that optimizes the minimum Levenshtein distance through explicit editing actions. RISE is able to pay attention to tokens that are related to conversational characteristics. To train RISE, we devise an Iterative Reinforce Training (IRT) algorithm with a Dynamic Programming based Sampling (DPS) process to improve exploration. Experimental results on two benchmark datasets show that RISE significantly outperforms state-of-the-art methods and generalizes well on unseen data.
Anthology ID:
2021.acl-long.438
Volume:
Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)
Month:
August
Year:
2021
Address:
Online
Venues:
ACL | IJCNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
5638–5650
Language:
URL:
https://aclanthology.org/2021.acl-long.438
DOI:
10.18653/v1/2021.acl-long.438
Bibkey:
Cite (ACL):
Zhongkun Liu, Pengjie Ren, Zhumin Chen, Zhaochun Ren, Maarten de Rijke, and Ming Zhou. 2021. Learning to Ask Conversational Questions by Optimizing Levenshtein Distance. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 5638–5650, Online. Association for Computational Linguistics.
Cite (Informal):
Learning to Ask Conversational Questions by Optimizing Levenshtein Distance (Liu et al., ACL 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.acl-long.438.pdf
Code
 LZKSKY/CaSE_RISE
Data
CANARD