Audience-Centric Natural Language Generation via Style Infusion

Samraj Moorjani, Adit Krishnan, Hari Sundaram, Ewa Maslowska, Aravind Sankar


Abstract
Adopting contextually appropriate, audience-tailored linguistic styles is critical to the success of user-centric language generation systems (e.g., chatbots, computer-aided writing, dialog systems). While existing approaches demonstrate text style transfer (TST) with large volumes of parallel or non-parallel data, we argue that grounding style on audience-independent external factors is innately limiting for two reasons. First, it is difficult to collect large volumes of audience-specific stylistic data. Second, some stylistic objectives (e.g., persuasiveness, memorability, empathy) are hard to define without audience feedback. In this paper, we propose the novel task of style infusion - infusing the stylistic preferences of audiences in pretrained language generation models. Since humans are better at pairwise comparisons than direct scoring - i.e., is Sample-A more persuasive/polite/empathic than Sample-B - we leverage limited pairwise human judgments to bootstrap a style analysis model and augment our seed set of judgments. We then infuse the learned textual style in a GPT-2 based text generator while balancing fluency and style adoption. With quantitative and qualitative assessments, we show that our infusion approach can generate compelling stylized examples with generic text prompts. We make the anonymized code and data accessible.
Anthology ID:
2022.findings-emnlp.138
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2022
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates
Editors:
Yoav Goldberg, Zornitsa Kozareva, Yue Zhang
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1919–1932
Language:
URL:
https://aclanthology.org/2022.findings-emnlp.138
DOI:
10.18653/v1/2022.findings-emnlp.138
Bibkey:
Cite (ACL):
Samraj Moorjani, Adit Krishnan, Hari Sundaram, Ewa Maslowska, and Aravind Sankar. 2022. Audience-Centric Natural Language Generation via Style Infusion. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 1919–1932, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
Cite (Informal):
Audience-Centric Natural Language Generation via Style Infusion (Moorjani et al., Findings 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.findings-emnlp.138.pdf
Software:
 2022.findings-emnlp.138.software.zip
Dataset:
 2022.findings-emnlp.138.dataset.zip
Video:
 https://aclanthology.org/2022.findings-emnlp.138.mp4