# Large language models for automated scholarly paper review: A survey

**Authors:** Zhenzhen Zhuang, Jiandong Chen, Hongfeng Xu, Yuwen Jiang, Jialiang Lin  
**arXiv:** [2501.10326](https://arxiv.org/abs/2501.10326)  
**Date:** 2025-01-17  
**Summary page:** https://www.summarizepaper.com/en/paper/2501.10326v1/

## Key points

- Large language models (LLMs) have revolutionized various aspects of human society
- Integration of LLMs into academic publications has impacted the peer review process, leading to automated scholarly paper review (ASPR)
- ASPR coexists alongside traditional peer review mechanisms
- Survey paper titled "Large language models for automated scholarly paper review: A survey" explores transformative potential of LLMs for ASPR and addresses new challenges
- Overview of ASPR implementation, identification of commonly used LLMs, overcoming technological bottlenecks, emergence of novel methods and systems
- Analysis of performance and issues associated with using LLMs in ASPR, attitudes and reactions of publishers and academia towards adopting ASPR
- Discussion on challenges researchers face in developing LLMs tailored for ASPR applications
- Valuable reference for researchers seeking to understand and contribute to progress in ASPR implementation

## Summary

Large language models (LLMs) have revolutionized various aspects of human society, with academia playing a crucial role in both shaping and being influenced by these advancements. The integration of LLMs into academic publications has notably impacted the peer review process, leading to the emergence of automated scholarly paper review (ASPR). This concept, initially proposed in previous research, has now evolved into a phase where ASPR coexists alongside traditional peer review mechanisms. In this survey paper titled "Large language models for automated scholarly paper review: A survey," authors Zhenzhen Zhuang, Jiandong Chen, Hongfeng Xu, Yuwen Jiang, and Jialiang Lin delve into the transformative potential of LLMs for ASPR while also addressing the new challenges they present. The study aims to provide a comprehensive overview of ASPR in the era of LLMs by exploring various facets of its implementation. The survey begins by identifying which specific LLMs are commonly utilized for conducting ASPR. Subsequently, the authors examine how technological bottlenecks related to ASPR have been overcome through the incorporation of LLM technology. They also investigate novel methods, datasets, source code, and online systems that have emerged alongside LLMs for ASPR applications. Furthermore, the paper delves into an analysis of the performance and issues associated with using LLMs in ASPR. It also explores the attitudes and reactions of publishers and academia towards adopting ASPR as a part of their manuscript review processes. Additionally, the study discusses the challenges that researchers face in developing LLMs tailored for ASPR applications. Overall,this survey serves as a valuable reference for researchers seeking to understand and contribute to the ongoing progress of ASPR implementation. By shedding light on the current landscape of LLM-driven scholarly paper reviews,it aims to inspire further advancements in this field and facilitate smoother integration of automated processes within academic publishing practices.