Towards Attribute-Entangled Controllable Text Generation: A Pilot Study of Blessing Generation

Shulin Huang, Shirong Ma, Yinghui Li, Li Yangning, Shiyang Lin, Haitao Zheng, Ying Shen


Abstract
Controllable Text Generation (CTG) has obtained great success due to its fine-grained generation ability obtained by focusing on multiple attributes. However, most existing CTG researches overlook how to utilize the attribute entanglement to enhance the diversity of the controlled generated texts. Facing this dilemma, we focus on a novel CTG scenario, i.e., blessing generation which is challenging because high-quality blessing texts require CTG models to comprehensively consider the entanglement between multiple attributes (e.g., objects and occasions). To promote the research on blessing generation, we present EBleT, a large-scale Entangled Blessing Text dataset containing 293K English sentences annotated with multiple attributes. Furthermore, we propose novel evaluation metrics to measure the quality of the blessing texts generated by the baseline models we designed. Our study opens a new research direction for controllable text generation and enables the development of attribute-entangled CTG models.
Anthology ID:
2022.gem-1.20
Volume:
Proceedings of the 2nd Workshop on Natural Language Generation, Evaluation, and Metrics (GEM)
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates (Hybrid)
Editors:
Antoine Bosselut, Khyathi Chandu, Kaustubh Dhole, Varun Gangal, Sebastian Gehrmann, Yacine Jernite, Jekaterina Novikova, Laura Perez-Beltrachini
Venue:
GEM
SIG:
SIGGEN
Publisher:
Association for Computational Linguistics
Note:
Pages:
235–247
Language:
URL:
https://aclanthology.org/2022.gem-1.20
DOI:
10.18653/v1/2022.gem-1.20
Bibkey:
Cite (ACL):
Shulin Huang, Shirong Ma, Yinghui Li, Li Yangning, Shiyang Lin, Haitao Zheng, and Ying Shen. 2022. Towards Attribute-Entangled Controllable Text Generation: A Pilot Study of Blessing Generation. In Proceedings of the 2nd Workshop on Natural Language Generation, Evaluation, and Metrics (GEM), pages 235–247, Abu Dhabi, United Arab Emirates (Hybrid). Association for Computational Linguistics.
Cite (Informal):
Towards Attribute-Entangled Controllable Text Generation: A Pilot Study of Blessing Generation (Huang et al., GEM 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.gem-1.20.pdf
Video:
 https://aclanthology.org/2022.gem-1.20.mp4