Learning to Model Editing Processes

Machel Reid, Graham Neubig


Abstract
Most existing sequence generation models produce outputs in one pass, usually left-to-right. However, this is in contrast with a more natural approach that humans use in generating content; iterative refinement and editing. Recent work has introduced edit-based models for various tasks (such as neural machine translation and text style transfer), but these generally model a single edit step. In this work, we propose modeling editing processes, modeling the whole process of iteratively generating sequences. We form a conceptual framework to describe the likelihood of multi-step edits, and describe neural models that can learn a generative model of sequences based on these multistep edits. We introduce baseline results and metrics on this task, finding that modeling editing processes improves performance on a variety of axes on both our proposed task and related downstream tasks compared to previous single-step models of edits.
Anthology ID:
2022.findings-emnlp.280
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:
3822–3832
Language:
URL:
https://aclanthology.org/2022.findings-emnlp.280
DOI:
10.18653/v1/2022.findings-emnlp.280
Bibkey:
Cite (ACL):
Machel Reid and Graham Neubig. 2022. Learning to Model Editing Processes. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 3822–3832, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
Cite (Informal):
Learning to Model Editing Processes (Reid & Neubig, Findings 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.findings-emnlp.280.pdf