Structure-to-Text Generation with Self-Training, Acceptability Classifiers and Context-Conditioning for the GEM Shared Task

Shreyan Bakshi, Soumya Batra, Peyman Heidari, Ankit Arun, Shashank Jain, Michael White


Abstract
We explore the use of self-training and acceptability classifiers with pre-trained models for natural language generation in structure-to-text settings using three GEM datasets (E2E, WebNLG-en, Schema-Guided Dialog). With the Schema-Guided Dialog dataset, we also experiment with including multiple turns of context in the input. We find that self-training with reconstruction matching along with acceptability classifier filtering can improve semantic correctness, though gains are limited in the full-data setting. With context-conditioning, we find that including multiple turns in the context encourages the model to align with the user’s word and phrasing choices as well as to generate more self-consistent responses. In future versions of the GEM challenge, we encourage the inclusion of few-shot tracks to encourage research on data efficiency.
Anthology ID:
2021.gem-1.12
Volume:
Proceedings of the 1st Workshop on Natural Language Generation, Evaluation, and Metrics (GEM 2021)
Month:
August
Year:
2021
Address:
Online
Venues:
ACL | GEM | IJCNLP
SIG:
SIGGEN
Publisher:
Association for Computational Linguistics
Note:
Pages:
136–147
Language:
URL:
https://aclanthology.org/2021.gem-1.12
DOI:
10.18653/v1/2021.gem-1.12
Bibkey:
Copy Citation:
PDF:
https://aclanthology.org/2021.gem-1.12.pdf