Improving Disentangled Text Representation Learning with Information-Theoretic Guidance

Pengyu Cheng, Martin Renqiang Min, Dinghan Shen, Christopher Malon, Yizhe Zhang, Yitong Li, Lawrence Carin


Abstract
Learning disentangled representations of natural language is essential for many NLP tasks, e.g., conditional text generation, style transfer, personalized dialogue systems, etc. Similar problems have been studied extensively for other forms of data, such as images and videos. However, the discrete nature of natural language makes the disentangling of textual representations more challenging (e.g., the manipulation over the data space cannot be easily achieved). Inspired by information theory, we propose a novel method that effectively manifests disentangled representations of text, without any supervision on semantics. A new mutual information upper bound is derived and leveraged to measure dependence between style and content. By minimizing this upper bound, the proposed method induces style and content embeddings into two independent low-dimensional spaces. Experiments on both conditional text generation and text-style transfer demonstrate the high quality of our disentangled representation in terms of content and style preservation.
Anthology ID:
2020.acl-main.673
Volume:
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics
Month:
July
Year:
2020
Address:
Online
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
7530–7541
Language:
URL:
https://aclanthology.org/2020.acl-main.673
DOI:
10.18653/v1/2020.acl-main.673
Bibkey:
Cite (ACL):
Pengyu Cheng, Martin Renqiang Min, Dinghan Shen, Christopher Malon, Yizhe Zhang, Yitong Li, and Lawrence Carin. 2020. Improving Disentangled Text Representation Learning with Information-Theoretic Guidance. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 7530–7541, Online. Association for Computational Linguistics.
Cite (Informal):
Improving Disentangled Text Representation Learning with Information-Theoretic Guidance (Cheng et al., ACL 2020)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2020.acl-main.673.pdf
Video:
 http://slideslive.com/38929080