Learning to Model Editing Processes

By Machel Reid, Graham Neubig. 2022

38 citations recorded by the source corpus.

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.

Subfields: Artificial Intelligence, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Natural Language Processing

Citations in the corpus

Self-Refine: Iterative Refinement with Self-Feedback (2023, background): “In such scenarios, modern LLMs may produce an intelligible output, but iterative refinement is necessary to ensure that all aspects of the task are met and the desired quality is achieved (Reid and Neubig, 2022; Schick et al., 2022a; Welleck et al., 2022).”
PEER: A Collaborative Language Model (2022, methodology): “Similar to prior work on text editing (Faltings et al., 2021; Reid and Neubig, 2022), our first step in overcoming this issue is turning to Wikipedia – a single source that comes close to fulfilling all our needs: It provides a full edit history including comments on a diverse set of topics, is larg”
Generating Sequences by Learning to Self-Correct (2022, background): “Self-correction relates to recent works on editing text, including modeling Wikipedia edits (Reid & Neubig, 2022; Faltings et al., 2021; Schick et al., 2022), which relies on supervised edits, unsupervised methods (Miao et al., 2019; Liu et al., 2020) that perturb sequences with simple operations…”
MINT: Evaluating LLMs in Multi-turn Interaction with Tools and Language Feedback (2023, unknown): “5 ( 16k”