Embedded Files
Fangping Lan, Qi Zhang, Eduard C. Dragut
Temple University, DM Lab
Overview
Overview
To our knowledge, this is the first survey that synthesizes text revision research through the lens of edit intentions, providing a unified view of datasets, taxonomies, identification methods, and applications. We review prior work across the full revision workflow, including revision corpus construction, edit intention taxonomy design, and edit intention identification. We further categorize representative datasets and methods, summarize downstream applications such as writing assistance and document edit summarization, and highlight key open research directions.
Contributions:
We consolidate terminology and organize prior work across (1) the full revision workflow, (2) datasets and corpus construction, (3) EITs, (4) identification methods, (5) applications, and (6) open challenges, into a unified six-dimensional view.
We introduce a lineage-aware perspective to characterize how EITs evolve across domains and granularities, and summarize common operations such as merging, splitting, and refinement.
We review edit intention identification approaches spanning manual annotation, crowdsourcing, neural models, and LLM-based methods, and connect them to downstream uses including writing assistance, revision behavior analysis, and document edit summarization, highlighting methodological trade-offs and evaluation pitfalls.
Citation
Citation
Page updated
Google Sites
Report abuse