Back to search

Georgia Institute of Technology

Studying Text Revision in Scientific Writing

Abstract

dc:description.abstract

Writing is essential for sharing scientific discoveries, and researchers devote significant effort to revising their papers to improve writing quality and incorporate new findings. The revision process encodes valuable knowledge, including logical and structural improvements at the document level and stylistic and grammatical refinements at the sentence and word levels. This dissertation presents a complete computational framework for extracting text revisions across different granularity and analyzing edits made for different purposes. The first contribution is developing the state-of-the-art methods for monolingual sentence alignment. We propose a neural CRF model that captures sequential dependencies and semantic similarity between sentences in parallel documents. The proposed approach outperforms previous methods by a large margin, and enables the creation of high-quality text simplification datasets, such as Newsela-Auto and Wiki-Auto. Next, to study fine-grained editing operations, we design a novel neural semi-Markov CRF alignment model for monolingual word alignment. This model unifies word and phrase alignments using variable-length spans and achieves state-of-the-art performance on both in-domain and out-of-domain evaluations. It also demonstrates utility in downstream tasks, such as automatic text simplification and sentence pair classification tasks. We further present arXivEdits, a dataset containing human-annotated sentence alignments, and fine-grained span-level edits across multiple versions of 751 research papers. Enabled by this corpus, we perform a detailed analysis of revision strategies in scientific writing, revealing common practices researchers use to improve their papers. Finally, this dissertation explores the human revision from a readability perspective through MedReadMe, a new dataset consisting of sentence-level readability ratings and complex span annotations for 4,520 medical sentences. This dataset supports fine-grained readability analysis and the evaluation of state-of-the-art readability metrics. By incorporating novel features, we significantly improve their correlation with human judgments.

Degree

thesis:*
Level thesis:degree_level
Doctoral
Department dc:contributor.department
Interactive Computing
Grantor dc:publisher
Georgia Institute of Technology
Year dc:date.issued
2025

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Jiang, Chao
Advisor dc:contributor.advisor
  • Xu, Wei
Committee members dc:contributor.committeemember
  • Ritter, Alan
  • Goyal, Kartik
  • Peng, Nanyun
  • Li, Cheng

Subjects

dc:subject × 2

Rights

Language dc:language.iso
en_US

Identifiers

dc:identifier.*
Handle dc:identifier.uri
https://hdl.handle.net/1853/77806

Chain of custody

source
Harvested from
Georgia Tech
Base URL
repository.gatech.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Jiang, Chao. Studying Text Revision in Scientific Writing. Doctoral thesis, Georgia Institute of Technology, 2025. https://hdl.handle.net/1853/77806