TWT: Table with Written Text for Controlled Data-to-Text Generation

Tongliang Li, Lei Fang, Jian-Guang Lou, Zhoujun Li


Abstract
Large pre-trained neural models have recently shown remarkable progress in text generation. In this paper, we propose to generate text conditioned on the structured data (table) and a prefix (the written text) by leveraging the pre-trained models. We present a new data-to-text dataset, Table with Written Text (TWT), by repurposing two existing datasets: ToTTo and TabFact. TWT contains both factual and logical statements that are faithful to the structured data, aiming to serve as a useful benchmark for controlled text generation. Compared with existing data-to-text task settings, TWT is more intuitive, the prefix (usually provided by the user) controls the topic of the generated text. Existing methods usually output hallucinated text that is not faithful on TWT. Therefore, we design a novel approach with table-aware attention visibility and copy mechanism over the table. Experimental results show that our approach outperforms state-of-the-art methods under both automatic and human evaluation metrics.
Anthology ID:
2021.findings-emnlp.107
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2021
Month:
November
Year:
2021
Address:
Punta Cana, Dominican Republic
Editors:
Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih
Venue:
Findings
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
1244–1254
Language:
URL:
https://aclanthology.org/2021.findings-emnlp.107
DOI:
10.18653/v1/2021.findings-emnlp.107
Bibkey:
Cite (ACL):
Tongliang Li, Lei Fang, Jian-Guang Lou, and Zhoujun Li. 2021. TWT: Table with Written Text for Controlled Data-to-Text Generation. In Findings of the Association for Computational Linguistics: EMNLP 2021, pages 1244–1254, Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
TWT: Table with Written Text for Controlled Data-to-Text Generation (Li et al., Findings 2021)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-3/2021.findings-emnlp.107.pdf
Video:
 https://preview.aclanthology.org/nschneid-patch-3/2021.findings-emnlp.107.mp4
Data
RotoWireTabFactWikiBio