---
title: "Punishing poor reviewers at CVPR"
date: 2025-03-04
tags:
    - machine learning
    - peer review
    - opinion
has_math: false
---

This year [CVPR](https://cvpr.thecvf.com/) pledged to make all authors
participate in peer review, and reject papers from authors who wrote
low-quality reviews.[^ref] Last week they [confirmed on
Twitter](https://x.com/CVPR/status/1894853624200863958) that they followed
through with this and rejected 19 papers. Presumably this is a tiny fraction of
the overall papers submitted, but I hope this is an effective deterrent for
future authors. At the very least, I'm glad that _some_ major conference tried
something like this!

[^ref]: <https://cvpr.thecvf.com/Conferences/2025/CVPRChanges>

---

**Update 2025-06-25**: NeurIPS is trying something similar now, see
<https://blog.neurips.cc/2025/05/02/responsible-reviewing-initiative-for-neurips-2025/>.
In particular:

- Reviewers who are also authors _will not be able to see their own papers'
  reviews until they submit their own reviews_. This also extends to their
  co-authors.
- Once again, authors who are alos reviewers can have their own papers
  desk-rejected if they submit poor reviews.

Overall this seems like a good initiative, but I hope this doesn't cause people
to simply not review at all...