README.md

---
license: apache-2.0
tags:
- Automated Peer Reviewing
- SFT
datasets:
- ECNU-SEA/SEA_data
---

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

Paper Link: https://arxiv.org/abs/2407.12857

Project Page: https://ecnu-sea.github.io/


## 🔥 News

- 🔥🔥🔥 SEA is accepted by EMNLP2024 !
- 🔥🔥🔥 We have made SEA series models (7B) public !

## Model Description
⚠️ **_This is the SEA-S model for content standardization, and the review model SEA-E can be found [here](https://huggingface.co/ECNU-SEA/SEA-E)._**

The SEA-S model aims to integrate all reviews for each paper into one to eliminate redundancy and errors, focusing on the major advantages and disadvantages of the paper. Specifically, we first utilize GPT-4 to integrate multiple reviews of a paper into one (From [ECNU-SEA/SEA_data](https://huggingface.co/datasets/ECNU-SEA/SEA_data)) that is in a unified format and criterion with constructive contents, and form an instruction dataset for SFT. After that, we fine-tune [Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) to distill the knowledge of GPT-4. Therefore, SEA-S provides a novel paradigm for integrating peer review data in an unified format across various conferences.

```bibtex
@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}
}
```