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

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

demo.mp4

🔥 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

1. Clone the GitHub Repository:

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

2. Set Up Python Environment:

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

3. Install SEA Dependencies:

   cd SEA
   pip install -r requirements.txt

4. 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 web_ui/download_model.py

5. 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 web_ui/run_webui.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}
}

📬 Contact

If you have any inquiries, suggestions, or wish to contact us for any reason, we warmly invite you to email us at [email protected].

💐 Acknowledgments

Thanks to Nougat and LLaMA-Factory for their foundational contributions to this repository.

⭐ Star History