Learning to Model Editing Processes

Machel Reid, Graham Neubig

Keywords: Edit, Representation Learning, Source-code, natural language editing

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. In this work, we propose to model 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 multi-step 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.

One-sentence Summary: We propose a new task, edit modeling to model document likelihood by way of its revisions and their respective changes.

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2205.12374/code)