SEA

Python

[EMNLP 2024 Findings] SEA is an automated paper review framework capable of generating comprehensive and high-quality review feedback with high consistency for papers, thereby assisting researchers in improving the quality of their work.

Last updated Jul 2, 2026

Stars

Forks

Issues

Stars/day

Attention Score

View on GitHub ↗ Open in github.dev

Topics

#automated-peer-reviewing #dataset #domain-llm #llm #natural-language-processing #peer-review #sft

Language breakdown

Python 99.1%

Shell 0.9%

▸ Files click to expand

README

Automated Peer Reviewing in Paper SEA: Standardization, Evaluation, and Analysis

Unsure about the shortcomings in your work? Here’s what you can do....

https://github.com/ecnu-sea/sea/assets/52284163/2473418b-be94-4691-96d8-7ba79ab4690b

🔥 News

2024.09: 🎉 SEA is accepted by EMNLP2024 ! 🥳🥳🥳🥂🥂🥂
2024.06: 🎉 We have made SEA series models and dataset public !

❓ What is SEA

SEA is a novel framework for automated paper reviewing based on three modules: Standardization, Evaluation, and Analysis. SEA is capable of generating comprehensive and high-quality review feedback with high consistency for papers, thereby assisting researchers in improving the quality of their work.

⚡️ Quickstart

Clone the GitHub Repository:

git clone https://github.com/ecnu-sea/SEA.git

Set Up Python Environment:

conda create -n sea python=3.10 -y
   conda activate sea

Install SEA Dependencies:

cd SEA
   pip install -r requirements.txt

Download SEA-E model:

You can download the SEA-E model from Hugging Face to a local path by yourself, or you can run the following download script code:

python webui/downloadmodel.py

Now you are ready to have fun:

Note that you can change the hyper parameters in run_webui.sh.

cd web_ui
   bash run_webui.sh

Tips: You can set the model path downloaded from Hugging Face in the webui/runwebui.sh file.

🛡 Disclaimer

It must be underscored that the primary objective of SEA is to provide informative reviews for authors to furnish authors with insightful critiques aimed at refining their works, rather than directly influencing decisions regarding the acceptance or rejection of the papers. Commercial use is not allowed., and we have emphasized this point in the supplementary clauses of the model's license.

🔎 Citation

@inproceedings{yu2024automated,
  title={Automated Peer Reviewing in Paper SEA: Standardization, Evaluation, and Analysis},
  author={Yu, Jianxiang and Ding, Zichen and Tan, Jiaqi and Luo, Kangyang and Weng, Zhenmin and Gong, Chenghua and Zeng, Long and Cui, RenJing and Han, Chengcheng and Sun, Qiushi and others},
  booktitle={Findings of the Association for Computational Linguistics: EMNLP 2024},
  pages={10164--10184},
  year={2024}
}