[P] Bird's-eye views of conference proceedings

In December, I [shared](https://www.reddit.com/r/MachineLearning/comments/raa9y7/p_an_arxivsanitylike_view_of_neurips_2021_papers/) a link to a project I created to get a quick overview of top NeurIPS 2021 papers. The [website](https://tanelp.github.io/neurips2021/) ordered all accepted papers by average review score and showed an 8-page overview thumbnail for each paper in addition to abstract, link to code, and other meta-data available from OpenReview.

I recently wanted to have the same overview for ICLR 2022. I processed the papers and uploaded the overview to [https://www.confviews.com/iclr2022/](https://www.confviews.com/iclr2022/).

https://preview.redd.it/jz4usfk2e7n81.png?width=1004&format=png&auto=webp&s=b7493d4d338934f1346589a1596a960c569064d9

I am now using a …

conference machinelearning