ICLR open review dataset
1.ICLR data is a dataset of scientific peer reviews available to help researchers study this important artifact.The dataset consists of over 5K paper informatiron and the corresponding accept/reject decisions in top-tier venues from ICLR conference, as well as over 20K textual peer reviews written by reviewer for a review of the papers.For details of data introduction and format,please see ./data/README.md
2.This experiment mainly uses three parts of data, including ICLR data, arXiv data and cite data. Other data are extracted or spliced from these data.
3.Our experiments mainly include data acquisition, data analysis, clustering analysis, correlation analysis, emotional analysis.For details, please see ./code/README.md
pip install -r requirements.txt
pip install tensorflow==1.15.0
1.We thank the openreview.net, arxiv.org for their commitment to promoting transparency and openness in scientific communication.
2.We used the ELECTRA pre-training model.https://github.com/google-research/electra