Improving User Controlled Table-To-Text Generation Robustness

Hanxu Hu, Yunqing Liu, Zhongyi Yu, Laura Perez-Beltrachini


Abstract
In this work we study user controlled table-to-text generation where users explore the content in a table by selecting cells and reading a natural language description thereof automatically produce by a natural language generator. Such generation models usually learn from carefully selected cell combinations (clean cell selections); however, in practice users may select unexpected, redundant, or incoherent cell combinations (noisy cell selections). In experiments, we find that models perform well on test sets coming from the same distribution as the train data but their performance drops when evaluated on realistic noisy user inputs. We propose a fine-tuning regime with additional user-simulated noisy cell selections. Models fine-tuned with the proposed regime gain 4.85 BLEU points on user noisy test cases and 1.4 on clean test cases; and achieve comparable state-of-the-art performance on the ToTTo dataset.
Anthology ID:
2023.findings-eacl.175
Volume:
Findings of the Association for Computational Linguistics: EACL 2023
Month:
May
Year:
2023
Address:
Dubrovnik, Croatia
Editors:
Andreas Vlachos, Isabelle Augenstein
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2317–2324
Language:
URL:
https://aclanthology.org/2023.findings-eacl.175
DOI:
10.18653/v1/2023.findings-eacl.175
Bibkey:
Cite (ACL):
Hanxu Hu, Yunqing Liu, Zhongyi Yu, and Laura Perez-Beltrachini. 2023. Improving User Controlled Table-To-Text Generation Robustness. In Findings of the Association for Computational Linguistics: EACL 2023, pages 2317–2324, Dubrovnik, Croatia. Association for Computational Linguistics.
Cite (Informal):
Improving User Controlled Table-To-Text Generation Robustness (Hu et al., Findings 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.findings-eacl.175.pdf
Video:
 https://aclanthology.org/2023.findings-eacl.175.mp4