Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

From Lengthy to Lucid: A Systematic Literature Review on NLP Techniques for Taming Long Sentences

Tatiana Passali, Efstathios Chatzikyriakidis, Stelios Andreadis, Thanos G. Stavropoulos, Anastasia Matonaki, Anestis Fachantidis, Grigorios Tsoumakas

TL;DR

The paper surveys NLP techniques for taming long sentences through sentence compression and sentence splitting, emphasizing a PRISMA-guided methodology, datasets, metrics, and comparative analyses. It chronicles a shift from statistical and optimization-based methods to deep learning and Transformer architectures, while noting gaps in weakly/self-supervised and LLM-based approaches and the limited use of LLMs to date. Key contributions include a comprehensive taxonomy, systematic comparisons on standard datasets, and a discussion of challenges such as meaning preservation, interpretability, and resource demands. The work highlights practical implications for readability, accessibility, and downstream NLP tasks, and outlines future directions to broaden data-efficient and cross-lingual capabilities.

Abstract

Long sentences have been a persistent issue in written communication for many years since they make it challenging for readers to grasp the main points or follow the initial intention of the writer. This survey, conducted using the PRISMA guidelines, systematically reviews two main strategies for addressing the issue of long sentences: a) sentence compression and b) sentence splitting. An increased trend of interest in this area has been observed since 2005, with significant growth after 2017. Current research is dominated by supervised approaches for both sentence compression and splitting. Yet, there is a considerable gap in weakly and self-supervised techniques, suggesting an opportunity for further research, especially in domains with limited data. We also observe that despite their potential, Large Language Models (LLMs) have not yet been widely explored in this area. In this survey, we categorize and group the most representative methods into a comprehensive taxonomy. We also conduct a comparative evaluation analysis of these methods on common sentence compression and splitting datasets. Finally, we discuss the challenges and limitations of current methods, providing valuable insights for future research directions. This survey is meant to serve as a comprehensive resource for addressing the complexities of long sentences. We aim to enable researchers to make further advancements in the field until long sentences are no longer a barrier to effective communication.

From Lengthy to Lucid: A Systematic Literature Review on NLP Techniques for Taming Long Sentences

TL;DR

The paper surveys NLP techniques for taming long sentences through sentence compression and sentence splitting, emphasizing a PRISMA-guided methodology, datasets, metrics, and comparative analyses. It chronicles a shift from statistical and optimization-based methods to deep learning and Transformer architectures, while noting gaps in weakly/self-supervised and LLM-based approaches and the limited use of LLMs to date. Key contributions include a comprehensive taxonomy, systematic comparisons on standard datasets, and a discussion of challenges such as meaning preservation, interpretability, and resource demands. The work highlights practical implications for readability, accessibility, and downstream NLP tasks, and outlines future directions to broaden data-efficient and cross-lingual capabilities.

Abstract

Long sentences have been a persistent issue in written communication for many years since they make it challenging for readers to grasp the main points or follow the initial intention of the writer. This survey, conducted using the PRISMA guidelines, systematically reviews two main strategies for addressing the issue of long sentences: a) sentence compression and b) sentence splitting. An increased trend of interest in this area has been observed since 2005, with significant growth after 2017. Current research is dominated by supervised approaches for both sentence compression and splitting. Yet, there is a considerable gap in weakly and self-supervised techniques, suggesting an opportunity for further research, especially in domains with limited data. We also observe that despite their potential, Large Language Models (LLMs) have not yet been widely explored in this area. In this survey, we categorize and group the most representative methods into a comprehensive taxonomy. We also conduct a comparative evaluation analysis of these methods on common sentence compression and splitting datasets. Finally, we discuss the challenges and limitations of current methods, providing valuable insights for future research directions. This survey is meant to serve as a comprehensive resource for addressing the complexities of long sentences. We aim to enable researchers to make further advancements in the field until long sentences are no longer a barrier to effective communication.
Paper Structure (37 sections, 5 equations, 5 figures, 7 tables)

This paper contains 37 sections, 5 equations, 5 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Survey Methodology
  3. Sentence Compression
  4. Methods
  5. Statistical Methods
  6. Optimization Techniques
  7. Dependency Parsing
  8. Deep Learning
  9. Reinforcement Learning
  10. Summary
  11. Datasets
  12. Extractive Datasets
  13. Abstractive Datasets
  14. Measures
  15. Compression Rate
  16. ...and 22 more sections

Figures (5)

  • Figure 1: Cumulative number of publications from 2000 to 2025 for sentence compression and sentence splitting, demonstrating the momentum and growing interest in the field.
  • Figure 2: PRISMA flow diagram of the study selection process.
  • Figure 3: Taxonomy of sentence compression methods
  • Figure 4: Cumulative number of publications per year from 2000 to 2025 for sentence compression along with landmarks of key technology adoption.
  • Figure 5: Cumulative number of publications per year from 2005 to 2025 for sentence splitting along with landmarks of key technology adoption.