# A Systematic Survey of Text Summarization: From Statistical Methods to Large Language Models

**Authors:** Haopeng Zhang, Philip S. Yu, Jiawei Zhang  
**arXiv:** [2406.11289](https://arxiv.org/abs/2406.11289)  
**Date:** 2024-06-17  
**Summary page:** https://www.summarizepaper.com/en/paper/2406.11289v1/

## Key points

- Authors: Haopeng Zhang, Philip S. Yu, Jiawei Zhang
- Advancements in deep neural networks
- Pre-trained language models (PLMs)
- Recent large language models (LLMs)
- Overview of datasets, evaluation metrics, and summarization methods pre-dating LLM era
- Traditional statistical methods
- Deep learning approaches
- PLM fine-tuning techniques
- Recent advancements in benchmarking, modeling, and evaluating summarization within LLM era
- Synthesizing existing literature and offering a cohesive overview
- Highlighting research trends and addressing open challenges
- Proposing promising research directions in text summarization field

## Summary

In their paper titled "A Systematic Survey of Text Summarization: From Statistical Methods to Large Language Models," authors Haopeng Zhang, Philip S. Yu, and Jiawei Zhang delve into the evolution of text summarization research in light of advancements in deep neural networks, pre-trained language models (PLMs), and recent large language models (LLMs). The survey offers a comprehensive examination of the progress made in text summarization through the lens of these paradigm shifts. Divided into two main parts, the paper first provides a detailed overview of datasets, evaluation metrics, and summarization methods predating the LLM era. This includes discussions on traditional statistical methods, deep learning approaches, and PLM fine-tuning techniques. The second part focuses on recent advancements in benchmarking, modeling, and evaluating summarization within the LLM era. By synthesizing existing literature and offering a cohesive overview, the survey not only highlights research trends but also addresses open challenges while proposing promising research directions in the field of summarization. Ultimately, this work aims to guide researchers through the dynamic landscape of text summarization research by providing insights into its evolution and future prospects.